A Novel Glycolysis Evaluation System for Stratified Management and Individualized Treatment of OSCC Patients

A Novel Glycolysis Evaluation System for Stratified Management and Individualized Treatment of OSCC Patients | 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 Novel Glycolysis Evaluation System for Stratified Management and Individualized Treatment of OSCC Patients Ruolan Tan, Jiajia Ye, Zhihong Xu, Han Tang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8687369/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Oral squamous cell carcinoma (OSCC) is one of the most common malignant tumors in the head and neck region, characterized by high rates of metastasis and recurrence and resistance to traditional chemotherapy. We included 474 OSCC patients through the Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA). Then we applied weighted correlation network analysis (WGCNA) to identify glycolysis-related genes and constructed a 12-gene prognostic signature related to glycolysis. Subsequently, we used the model formula to divide the patients into two groups, and analyzed the differences in prognosis, energy metabolic patterns, gene expression, gene mutation, immunity, and therapeutic response between the two groups were analyzed to elucidate underlying mechanisms. Our analysis revealed patients with high-risk score exhibited poorer overall survival and a greater reliance on glycolysis for energy production. Additionally, significant differences were observed between the two groups in biological processes such as angiogenesis, epithelial cell migration, metalloendopeptidase activity, keratinization and humoral immune response. The High-risk group displayed lower infiltration levels of CD8 + T cells and higher TP53 mutation rates compared with the Low-risk group. Differences in drug resistance and sensitivity were also noted between the two groups. Collectively, Energy metabolic reprogramming influences the prognosis of OSCC patients, likely due to variations in gene expression, gene mutation, and immunocyte infiltration, particularly CD8 + T cells. Energy metabolic reprogramming Oral squamous cell carcinoma CD8+ T cell Cancer hallmarks Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Oral squamous cell carcinoma (OSCC) is a highly recurrent form of cancer arising from the mucosal lining of the oral cavity, characterized by high rates of metastasis, recurrence, and resistance to traditional chemotherapy [ 1 ]. It is one of the most common malignancies of the head and neck region with ~ 373,000 new patients diagnosed and ~ 199,000 deaths reported in 2019 [ 2 ]. Various etiological factors, including smoking, betel nut chewing, immunodeficiency, and alcohol consumption, contribute to mutational changes, which further result in the appearance of OSCC. Conventional therapies of OSCC are surgery, chemotherapy, radiotherapy or a combination of these modalities. Despite advances in cancer diagnosis and treatment, the overall survival of OSCC patients has not improved substantially over the past four decades [ 3 ]. One-third of OSCC patients eventually develop life-threatening and incurable recurrent disease [ 4 ]. Due to patient heterogeneity, cancer metastasis, neoplasm recurrence, and drug resistance, tumors at similar stages respond very differently to the same treatment. Thus, if proper characteristics can be used to stratify patients for precision and personalized management plans, it will greatly improve the prognosis of OSCC patients [ 5 ]. In 2022, Hanahan proposed fourteen hallmarks in the development of tumors, emphasizing the role of reprogramming energy metabolism in promoting tumor progression [ 6 ]. The rapid and unbridled proliferation characteristic of tumor growth is an energy- and resource-consuming process, and thus metabolism is significantly altered during neoplastic transformation and tumor progression [ 7 ]. Under aerobic condition, normal cells produce energy primarily through mitochondria, and pyruvate produced by glycolysis entering the mitochondria for the subsequent tricarboxylic acid (TCA) cycle and oxidative phosphorylation (OXPHOS). in contrast, under anaerobic conditions, glycolysis is more advantageous, and pyruvate is primarily converted to lactate with little transfer to the mitochondria. This state is known as the Warburg effect or aerobic glycolysis, first described in the 1920s [ 8 ]. These alterations cause most cancers to induce unregulated glucose fermentation pathways for energy and to fuel growth, a factor underlying the use of 18 F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) as an important diagnostic tool for oncologists [ 7 ]. Based on this characteristic, Abd El-Hafez et al. [ 9 ] calculated a total lesion glycolysis (TLG) of each OSCC patient from PET/CT in a prospective study and observed higher rates of distant metastases and worse prognosis in patients with high TLG. Furthermore, Ryu et al. [ 5 ] retrospectively evaluated the ability of TLG to stratify the likelihood of survival and predict occult metastasis in OSCC, finding that patients with high TLG had inferior outcomes, providing guidance for the formulation of treatment and follow-up strategies. However, the absence of a real quantized gold standard and high price limit the application and promotion of PET/CT-based prognostic assessment. Despite increasing evidence demonstrating that glycolysis can promote the growth, invasion and migration of tumor cells [ 10 , 11 ], the effect of glycolysis on patient's long-term prognosis and the underlying mechanistic details pertinent to the causes and consequences of such metabolic phenotype remain unclear. In this study, we explored the influence of fourteen hallmarks of cancer on the overall survival of OSCC patients. Among them, glycolysis was identified as a risk factor for the prognosis of OSCC. Subsequently, glycolysis-related genes were detected, and a prognosis model was constructed based on these genes. According to the model, patients were separated into two groups, which exhibited different energy metabolic patterns, infiltration levels of immunocytes, gene expression and mutation, response to chemotherapeutic drugs, and prognosis, suggesting a potential approach to stratify and manage patients (the data analysis process is shown in Fig. 1 ). Moreover, the glycolysis-related model genes have research prospects as new biomarkers for regulating metabolic reprogramming and tumor progression, providing a reference for a new quantitative method of glycolysis. Materials and Methods Data acquisition and process A total of three OSCC public cohorts with comprehensive gene expression profiles and clinical information were included in this study. Two datasets (GSE41613 and GSE42743) from one study were acquired from Gene Expression Omnibus (GEO) database [ 12 ]. The raw CEL files were downloaded and integrated into a new cohort using robust multiarray method for data normalization and the Combat algorithm to eliminate batch effects. The combined dataset of 171 OSCC samples was used as a training set. The validation cohort consisted of 303 patients with whole RNA-sequencing data, age, gender, tumor stage, and overall survival information was collected from UCSC Xena ( http://xena.ucsc.edu ). Somatic mutation data, sorted in Mutation Annotation Format (MAF), were obtained from The Cancer Genome Atlas (TCGA, https://portal.gdc.cancer.gov ). Probe IDs were mapped to gene symbols according to the corresponding annotation file, and probes targeting genes in common were summarized by selecting those showing the highest mean of expression between all samples. Fourteen main gene sets of cancer-related hallmarks [ 6 , 13 ] were retrieved from Molecular Signatures Database (MSigDB, https://www.gsea-msigdb.org/gsea/msigdb/index.jsp ) and previously published literatures [ 14 ]. Statistical methods Construction and Evaluation of an Energy Metabolism-Related Prognosis Model The single-sample gene set enrichment analysis (ssGSEA) algorithm was used by R package “gsva” to calculate the levels of cancer-related hallmarks in each training sample. Multivariate Cox regression, realized by R package “survminer”, was performed to evaluate the prognosis value of the signatures. To further explore the key hallmark (glycolysis) associated genes, we carried out Weighted correlation network analysis (WGCNA) using the R package “WGCNA”. Next, the least absolute shrinkage and selection operator (LASSO) Cox regression analysis was applied using the R package “glmnet” to narrow down the variates, screen the most robust candidates, and build a glycolysis-related prognostic model. Additionally, Kaplan-Meier curve analysis was used to determine the value of the model for prognostic assessment. Differential, Functional Enrichment, Immune Infiltration and Mutation Analyses The R package “DESeq2” was used to identify differentially expressed genes (DEGs). Genes with a log 2 fold change (log 2 FC) > 1 or < -1 and an adjust p -value < 0.05 were identified as DEGs. Functional enrichment analysis was implemented in Cytoscape software 3.7.2. For mutation data, the R package “maftools” was used to investigate mutation sites and frequencies. To quantify the relative proportions of infiltrating immune cells from the gene expression profiles, the R package “CIBERSORT” was used. The putative abundance of immune cells was estimated using a reference set with 22 types of immune cell subtypes (LM22) with 1,000 permutations. Gene Set Enrichment Analysis (GSEA) and Assessment of Therapeutic Response The gene sets associated with drug resistance were acquired from MSigDB, and GSEA was evaluated using the R package “clusterProfiler”. Furthermore, the R package “oncoPredict” was applied to estimate the chemotherapeutic responses of patients. Results Glycolysis Identified as an Important Prognostic Risk Factor for Overall Survival in OSCC The prognostic value of the hallmarks was estimated by multivariate Cox regression analysis. Remarkably, glycolysis was the only significant risk factor for overall survival among various cancer-related hallmarks, with a HR = 10.445 and a p- value = 0.042 (Fig. 2 A). Subsequently, the heatmap exhibited close correlations between overall survival status and tumor stage with glycolysis (Fig. 2 B). Exploration of Glycolysis-Associated Candidate Genes WGCNA was performed with transcriptome profiling data and glycolysis ssGSEA scores to construct a scale-free co-expression network. With a soft threshold of 5, the genes were separated to 44 modules, and the lightgreen module displayed the strongest correlation with glycolysis, indicating it as the source module of candidate genes (Fig. 2 C). Furthermore, LASSO regression helped to narrow down the variates, and twelve genes were selected to construct a prognostic model (Fig. 3 A and 3 B). Among the 12 genes, three ( PIK3C2B , MEX3D , BMP2 ) with coefficients 0, Fig. 3 C). According to the formula, risk score = \(\:\sum\:_{i=1}^{n}{Coefficietn\:RNA}_{i}\:\times\:\:Expression\:{RNA}_{i}\) , the risk score of each patient was calculated and then normalized to Z-score. Patients with a risk Z-score > 0 were identified as High-risk group and others were classified into Low-risk group. The scatter diagram revealed the survival time and status of patients in different groups. The heatmap further exhibited the expression levels of the candidate genes in training and validation cohorts. As shown in Fig. 3 D and 3 G, most genes were significantly different between High- and Low-risk groups, except for PIK3C2B in the validation cohort. Kaplan-Meier curve analysis demonstrated that the Low-risk group had a better outcome compared with the High-risk group (Fig. 3 E and 3 H). Since this prognostic model was constructed based on glycolysis-related candidate genes, we further focused on the differences in energy metabolic patterns between the High- and Low-risk groups. As shown in Fig. 3 F and 3 I, the High-risk group may primarily generate energy through glycolysis, while the Low-risk group may tend to obtain energy through the TCA cycle and OXPHOS. Depicting of Energy Metabolic map with Differentially Expressed Genes To further elucidate the variational trend of related genes during glycolysis, the TCA cycle and OXPHOS, we drew an energy metabolic map. Genes with the same expression tendency in both GEO and TCGA cohorts and significant differences between the High-risk and Low-risk groups in the TCGA cohort were selected and exhibited in the map (Fig. 4 ). As shown in the map, genes regulating the conversion of pyruvate to acetyl-COA were down-regulated in the High-risk group, suggesting that less pyruvate produced by glycolysis was converted to acetyl-COA and entered the TCA cycle. Similarly, the relevant regulatory genes in each OXPHOS complex were down-regulated, indicating that less energy was generated through OXPHOS. Taken together, patients in the High-risk group may be less dependent on the TCA cycle and OXPHOS, further confirming our previous hypothesis about energy metabolic patterns in the High- and Low-risk groups. Comprehensive Analyses of Genomic Alterations between the High-risk group and Low-risk group Based on the previous analysis, OSCC patients were preliminarily divided into High-risk and Low-risk groups, and the potential differences in energy metabolic patterns and prognostic survival were observed between the two groups. Next, we analyzed gene expression and mutation differences between the two groups to further explore the underlying mechanisms. There were 264 genes up-regulated and 1,564 genes down-regulated in the High-risk group (Fig. 5 A). Functional enrichment analysis of the increased and decreased genes, respectively, revealed that the up-regulated genes were mainly enriched in angiogenesis, epithelial migration, and metallopeptidase activity etc., which were considered to induce or enhance tumor growth, invasion, and metastasis [ 15 ]. In contrast, the down-regulated genes mainly involved in immune response not conducive to tumor progression, such as epithelial development and keratinization (Fig. 5 B). Subsequently, CIBERSORT was used to investigate the different infiltration levels of immunocytes between the two groups, and immunocytes with more than 60% of samples scoring zero were eliminated. Most of the immune cells were significantly infiltrated at higher levels in the Low-risk group, including plasma cells, CD8 + T cells, follicular helper T cells, regulatory T cells (Tregs), and resting dendritic cells. Two immunocytes, resting memory CD4 + T cells and M0 macrophage, exhibited higher infiltration levels in High-risk group (Fig. 5 C). Additionally, Spearman’s correlation analysis revealed that the follicular helper T cells, Tregs and CD8 + T cells were the top three immune cells negatively correlated with risk Z-score, while M0 macrophage showed the strongest positive correlation with the risk Z-score. CD8 + T cells are crucial members of cytotoxic T cells that kill tumor cells, and their decreased number may lead to a weakened eliminating effect on tumor cells. Further analysis found that the High-risk group had a higher T cell exclusion potential of the tumor (Supplementary Fig. 1), indicating that more T cell exclusion may occur in the High-risk group, leading to a lower T cell infiltration level. There is evidence that gene mutations play a crucial role in tumor angiogenesis, invasion and migration [ 16 ]. Therefore, we further focused on the differences in mutations between the High-risk and Low-risk groups. Figure 5 D and 5 E showed the top 20 mutated genes in the two groups. TP53 , TTN , FAT1 , CDKN2A , NOTCH1 ranked in the top 5, with TP53 being the most frequently mutated gene in both groups. Fifteen identical mutated genes among the top 20 genes in both the High-risk and Low-risk groups were selected for further analysis. Higher mutation rates were observed in TP53 in the High-risk group compared with Low-risk group (Fig. 5 F). The lollipop plot showed the differences in TP53 mutation sites between the two groups (Fig. 5 G). To sum up, the significant differences in angiogenesis, cell migration, immune response and mutated genes may be related to differences in energy metabolism and survival between the High-risk group and Low-risk group. Energy Metabolic Pattern Associated with Therapeutic Response Management of OSCC often requires multimodality treatments, including surgery, radiation, and/or chemotherapy. Previous studies have expounded that mutations in TP53 are significantly associated with poor survival and tumor resistance to radiotherapy and chemotherapy in OSCC patients [ 17 ]. Furthermore, glycolysis and glycolytic enzymes have been regarded as pivotal factors promoting a drug-resistant phenotype [ 18 ]. Therefore, we speculated that the differences in TP53 mutation and energy metabolic patterns between the High-risk and Low-risk groups might cause different chemotherapeutic sensitivity to chemotherapy drugs in the two groups. Based on the altered gene sets of different drug treatments retrieved, GSEA predicted that resistance to doxorubicin, gefitinib, cisplatin and imatinib was significantly up-regulated, and resistance to gemcitabine, trabectedin and tamoxifen was down-regulated in the High-risk group, indicating a close correlation between drug resistance and energy metabolism pattern (Fig. 6 A). Furthermore, we focused on five drugs that have been used and/or studied in OSCC chemotherapy (Fig. 6 B). By estimating the IC50 value of each patient, we compared the chemotherapeutic sensitivity between the two groups. Patients in the High-risk group had higher IC50 values for doxorubicin and gefitinib and lower IC50 values for tamoxifen, consistent with the GSEA results. However, drug sensitivity to cisplatin exhibited no difference, and opposite results were perceived for two types of gemcitabine (Fig. 6 C). These results denoted that energy metabolic patterns have the potential to affect patients’ chemotherapeutic sensitivity to drugs, which may further result in the different prognosis. Discussion Expanding incidence of OSCC malignancies and late-stage presentation have made it global healthcare issues. Traditionally, the management of OSCC begins with initial staging using the TNM system, which, however, is insufficient for predicting disease prognosis [ 19 ]. This underscores the critical need for precise identification and stratification of these malignant lesions to maximize patient outcomes. Herein, we first validated the prognostic significance of glycolysis in OSCC (Fig. 2 A). Cancer cells often reprogram their glucose metabolic pathways to adapt to environmental challenges and support survival, proliferation, and metastasis [ 20 ]. Previous studies have suggested that glycolysis is the major glucose metabolic pathway in OSCC cell lines, with inhibited glycolysis activating intrinsic apoptosis and accelerated glycolysis promoting in vivo tumor development [ 11 ]. Thus, it is not surprising that the glycolysis adversely affects the prognosis of OSCC patients, as observed in our study. After establishing glycolysis as a prognosis risk factor for OSCC, we further investigated the genes associated glycolysis (Fig. 2 C). Based on these genes ( PIK3C2B , MEX3D , BMP2 , SLC20A1 , TMEM181 , WDR54 , SNAPC1 , S1PR5 , STC2 , ANXA5 , THBS1 , KLF7 ), we developed a glycolysis-related prognostic model (Fig. 3 A and 3 B). Using this model, patients were stratified into High- and Low-risk groups, and distinct outcomes and energy metabolic patterns (Fig. 3 E, 3 F, 3 H, and 3 I). Specifically, the High-risk groups had a poorer prognosis and relied primarily on glycolysis for energy metabolism. Previous studies have attempted to assess tumor glycolysis levels by PET/CT to predict patient prognosis [ 5 , 9 ]. However, the complexity and cost of this method limits its application. The calculation formula presented in this study offers a novel tool that is more convenient, more efficient, more affordable and more accessible for assessing glycolysis levels and predicting OSCC patient outcomes. It is well recognized that in tumor cells prefer glycolysis as their primary energy source, even in the presence of sufficient oxygen. This results in increased lactate production and reduced pyruvate entry into mitochondria, leading to decrease mitochondrial energy metabolism [ 21 ]. Our study also observed this phenomenon, with down-regulated expression of genes catalyzing pyruvate oxidation to acetyl-CoA in the High-risk group, reducing acetyl-CoA production (Fig. 4 ). As a crucial driver of the TCA cycle, acetyl-CoA is essential for sustaining TCA cycle activity. Reduced acetyl-CoA formation directly impacts subsequent mitochondrial energy metabolism [ 22 ]. With weakened TCA cycle activity, the amount of succinate entering OXPHOS decreased, and the expressions of OXPHOS complex I-V genes were down-regulated, further limiting energy production through OXPHOS. Despite generating less energy than OXPHOS, glycolysis remains the preferred way for tumor cell proliferation and growth due to its efficient glucose uptake and productivity [ 23 ]. These findings partially explain why the High-risk group, with high glycolysis and low OXPHOS as their primary energy metabolic pattern, had a worse prognosis than the Low-risk group. Furthermore, we found that the genes with higher expression in the High-risk group were enriched in biological processes conductive to tumor development, such as cell adhesion mediated by integrins, epithelial cell migration, metallopeptidase and metalloendopeptidase activity, and angiogenesis. Integrins are involved in nearly every step of cancer progression, supporting tumor invasion, acquisition of tumor stem cell properties, and drug resistance [ 24 ]. Moreover, extensive research has identified matrix metalloproteinases (MMPs) as the most prominent proteinases family associated with tumorigenesis. MMP activity can induce or enhance tumor survival, invasion and metastasis [ 15 ]. Consequently, the elevated expression of these associated genes suggests that tumors in patients with a High-risk score may exhibit faster growth and more aggressive behavior compared to the Low-risk group. On the other hand, the down-regulated genes are primarily involved in epithelial cell differentiation, keratinocyte differentiation, arachidonic acid monooxygenase activity, and humoral immune response. Dyskeratosis is a prognostic histopathological feature that impacts the survival outcomes of patients with OSCC. Recent researchers have observed that OSCC patients with minimal or absent keratinization tend to experience more recurrences, lower 5-year disease-free and disease-specific survival rates, and significantly higher lymph node metastasis compared to those with good or high keratinization [ 25 ]. On the other hand, host immunity represents the human body’s most crucial defensive mechanism. Immune cells within the tumor microenvironment (TME) can either eliminate cancerous cells or promote tumor growth by secreting anti-tumor cytokines, growth factors, or pro-inflammatory factors [ 26 ]. The TME’s hostile environment for anti-tumor immune cells is partly driven by a cascade of biochemical reactions leading to local acidity [ 27 ]. Preliminary evidence indicates that low pH can facilitate immune escape by both attenuating antitumor effectors and promoting the recruitment and activity of immunosuppressive pro-tumor immune cells [ 27 ]. Given our previous observations, we speculated that since the High-risk group relies primarily on glycolysis as their metabolic mode, more pyruvate is converted into lactate, altering the local pH and impacting immune cell function. Collectively, these findings further affirm that the High-risk group possesses more advantages in the tumor formation and development process. Given the previous results suggesting potential immune differences between the High-risk and -Low groups, we focused on immune cell infiltration levels. A significant decrease in CD8 + T cells was observed in High-risk group, with a negative correlation between CD8 + T cells and EMPS Z-score (Fig. 5 C). As a key member of tumor-infiltration T lymphocytes, CD8 + T cells selectively recognize and eliminate cancer cells by interacting with the cognate T cell receptor (TCR) and the HLA/antigen complex, while also secreting anti-tumor cytokines such as interferon (IFN)-γ, tumor necrosis factor (TNF)-α, interleukin (IL)-17, and IL-2 [ 26 ]. Withal, Shimizu et al. [ 28 ] enunciated that previously untreated OSCC patients with high tumor-infiltration CD8 + T cells had significantly better outcomes, emphasizing the strong correlation between CD8 + T cells and the prognosis of OSCC patients. Therefore, the lower infiltration level of CD8 + T cells may contribute to the poorer prognosis in High-risk group. Further analysis revealed that the decrease of CD8 + T cells may be related to an increase in T cell exclusion potential (Supplementary Fig. 1). By comparing the mutation differences, we noted that TP53 mutation were significantly higher in the High-risk group than in the Low-risk group (Fig. 5 F). TP53 mutations are common in human tumors and are linked to poor patient prognosis in many cancers. The wild-type p53 encoded by TP53 is considered an effective tumor suppressor that promotes cell cycle arrest and apoptosis, thereby inhibiting cell growth and oncogenic transformation [ 29 ]. However, mutations in TP53 lead to the loss p53’s tumor suppressive activities, allowing oncogene-expressing cells to proliferate unabated, directly contributing to cancer development [ 29 ]. Consistent with our observations, the High-risk group with a poorer prognosis showed a higher frequency of TP53 mutations, which may ultimately accelerate tumor progression. Overall, the combined effects of differences in energy metabolism pattern, gene expression, immune infiltration and mutation may explain the distinct prognosis between the two groups. Chemotherapy serves as a crucial adjuvant in the treatment of tumors. Research has demonstrated that a heightened glycolytic rate in tumor cells contributes to an increased resistance to chemotherapeutic agents [ 30 ]. Furthermore, mutations in the TP53 gene can lead to enhanced resistance to chemotherapy drugs [ 17 ]. Does the present research exhibit a similar trait? We conducted a further analysis of the therapeutic response to chemotherapy drugs in both the High-risk and Low groups. We selected five drugs that have been applied and/or investigated in the chemotherapy of OSCC and found that the High-risk group did not display heightened resistance to all chemotherapy drugs (Fig. 6 A and 6 C). For cisplatin, a commonly prescribed chemotherapeutic agent for OSCC, no significant increase in resistance or decrease in drug sensitivity was observed in the High-risk group. However, for doxorubicin and gefitinib, there may be an increase in resistance. Although these findings do not fully align with our initial hypotheses, they still offer valuable insights for the selection of chemotherapy management strategies for OSCC patients. Based on glycolysis-related genes, this study constructed a more convenient, more economical and more generalized glycolysis computational model and initially explored its ability to classify patients and predict patient outcomes. We discovered that patients who rely on glycolysis as their primary energy metabolic mode tend to have a poorer prognosis. This may be associated with gene expression levels, mutation status, as well as immune status. Additionally, we conducted a preliminary investigation into the potential causes of glycolysis affecting patients’ prognosis and aimed to propose a new target for intervening in tumor glycolysis. Simultaneously, we endeavor to introduce a novel method for evaluating tumor metabolic pattern and patient stratification, offering a fresh perspective for personalized patient treatment. Declarations Acknowledgement This work was supported by the National Natural Science Foundation of China (Grant No.82501120). We would like to thank all the study team members for their support and contribution in any form. Funding This work was supported by the National Natural Science Foundation of China (Grant No.82501120). Author Contributions Conceptualization, Methodology, Supervision: Zhihong Xu and Han Tang; Data curation, Investigation: Ruolan Tan and Han Tang; Formal analysis, Software, Validation, Visualization: Ruolan Tan and Jiajia Ye; Writing – original draft: Ruolan Tan and Jiajia Ye; Writing – review & editing: Zhihong Xu and Han Tang. All authors have read and approved the final manuscript. Data Availability The datasets generated and analyzed during the current study are available in UCSC Xena (http://xena.ucsc.edu) and Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo/), reference number [GSE41613 and GSE42743]. Competing interests The authors have no relevant financial or non-financial interests to disclose. Consent to participate N/A. Ethical approval N/A. References Chai AWY, Lim KP, Cheong SC. Translational genomics and recent advances in oral squamous cell carcinoma. Semin Cancer Biol. 2020;61:71–83. https://doi.org/10.1016/j.semcancer.2019.09.011 . Global Burden of Disease Cancer, Incidence C, et al. Mortality, Years of Life Lost, Years Lived With Disability, and Disability-Adjusted. JAMA Oncol. 2022;8(3):420–44. https://doi.org/10.1001/jamaoncol.2021.6987 . Life Years for 29 Cancer Groups From 2010 to 2019: A Systematic Analysis for the Global Burden of Disease Study 2019. Zini A, Czerninski R, Sgan-Cohen HD. Oral cancer over four decades: epidemiology, trends, histology, and survival by anatomical sites. J Oral Pathol Med. 2010;39(4):299–305. https://doi.org/10.1111/j.1600-0714.2009.00845.x . Ling Z, Cheng B, Tao X. Epithelial-to-mesenchymal transition in oral squamous cell carcinoma: Challenges and opportunities. Int J Cancer. 2021;148(7):1548–61. https://doi.org/10.1002/ijc.33352 . Ryu IS, et al. Prognostic significance of preoperative metabolic tumour volume and total lesion glycolysis measured by (18)F-FDG PET/CT in squamous cell carcinoma of the oral cavity. Eur J Nucl Med Mol Imaging. 2014;41(3):452–61. https://doi.org/10.1007/s00259-013-2571-z . .doi:. Hanahan D. Hallmarks of Cancer: New Dimensions. Cancer Discov. 2022;12(1):31–46. https://doi.org/10.1158/2159-8290.CD-21-1059 . Poff A, et al. Targeting the Warburg effect for cancer treatment: Ketogenic diets for management of glioma. Semin Cancer Biol. 2019;56:135–48. https://doi.org/10.1016/j.semcancer.2017.12.011 . Warburg O, Wind F, Negelein E. The Metabolism of Tumors in the Body. J Gen Physiol. 1927;8(6):519–30. https://doi.org/10.1085/jgp.8.6.519 . Abd El-Hafez YG, et al. Total lesion glycolysis: a possible new prognostic parameter in oral cavity squamous cell carcinoma. Oral Oncol. 2013;49(3):261–8. https://doi.org/10.1016/j.oraloncology.2012.09.005 . Gong X, Tang H, Yang K. PER1 suppresses glycolysis and cell proliferation in oral squamous cell carcinoma via the PER1/RACK1/PI3K signaling complex. Cell Death Dis. 2021;12(3):276. https://doi.org/10.1038/s41419-021-03563-5 . Li M, et al. Tanshinone IIA inhibits oral squamous cell carcinoma via reducing Akt-c-Myc signaling-mediated aerobic glycolysis. Cell Death Dis. 2020;11(5):381. https://doi.org/10.1038/s41419-020-2579-9 . Lohavanichbutr P, et al. A 13-gene signature prognostic of HPV-negative OSCC: discovery and external validation. Clin Cancer Res. 2013;19(5):1197–203. https://doi.org/10.1158/1078-0432.CCR-12-2647 . Shi R, et al. Identification and validation of hypoxia-derived gene signatures to predict clinical outcomes and therapeutic responses in stage I lung adenocarcinoma patients. Theranostics. 2021;11(10):5061–76. https://doi.org/10.7150/thno.56202 . Miranda A, et al. Cancer stemness, intratumoral heterogeneity, and immune response across cancers. Proc Natl Acad Sci U S A. 2019;116(18):9020–9. https://doi.org/10.1073/pnas.1818210116 . Kessenbrock K, Plaks V, Werb Z. Matrix metalloproteinases: regulators of the tumor microenvironment. Cell. 2010;141(1):52–67. https://doi.org/10.1016/j.cell.2010.03.015 . Sabapathy K, Lane DP. Therapeutic targeting of p53: all mutants are equal, but some mutants are more equal than others. Nat Rev Clin Oncol. 2018;15(1):13–30. https://doi.org/10.1038/nrclinonc.2017.151 . Lindemann A, et al. Targeting the DNA Damage Response in OSCC with TP53 Mutations. J Dent Res. 2018;97(6):635–44. https://doi.org/10.1177/0022034518759068 . Bhattacharya B, Mohd MF, Omar, Soong R. The Warburg effect and drug resistance. Br J Pharmacol. 2016;173(6):970–9. https://doi.org/10.1111/bph.13422 . Chakraborty D, et al. A facile gold nanoparticle-based ELISA system for detection of osteopontin in saliva: Towards oral cancer diagnostics. Clin Chim Acta. 2018;477:166–72. https://doi.org/10.1016/j.cca.2017.09.009 . Reinfeld BI, et al. The therapeutic implications of immunosuppressive tumor aerobic glycolysis. Cell Mol Immunol. 2022;19(1):46–58. https://doi.org/10.1038/s41423-021-00727-3 . DeBerardinis RJ, Chandel NS. Fundamentals of cancer metabolism. Sci Adv. 2016;2(5):e1600200. https://doi.org/10.1126/sciadv.1600200 . Martinez-Reyes I, Chandel NS. Mitochondrial TCA cycle metabolites control physiology and disease. Nat Commun. 2020;11(1):102. https://doi.org/10.1038/s41467-019-13668-3 . Ganapathy-Kanniappan S. Linking tumor glycolysis and immune evasion in cancer: Emerging concepts and therapeutic opportunities. Biochim Biophys Acta Rev Cancer. 2017;1868(1):212–20. https://doi.org/10.1016/j.bbcan.2017.04.002 . Hamidi H, Ivaska J. Every step of the way: integrins in cancer progression and metastasis. Nat Rev Cancer. 2018;18(9):533–48. https://doi.org/10.1038/s41568-018-0038-z . Wolfer S, Elstner S, Schultze-Mosgau S. Degree of Keratinization Is an Independent Prognostic Factor in Oral Squamous Cell Carcinoma. J Oral Maxillofac Surg. 2018;76(2):444–54. https://doi.org/10.1016/j.joms.2017.06.034 . Jiang X, et al. Role of the tumor microenvironment in PD-L1/PD-1-mediated tumor immune escape. Mol Cancer. 2019;18(1):10. https://doi.org/10.1186/s12943-018-0928-4 . Huber V, et al. Cancer acidity: An ultimate frontier of tumor immune escape and a novel target of immunomodulation. Semin Cancer Biol. 2017;43:74–89. https://doi.org/10.1016/j.semcancer.2017.03.001 . Shimizu S, et al. Tumor-infiltrating CD8(+) T-cell density is an independent prognostic marker for oral squamous cell carcinoma. Cancer Med. 2019;8(1):80–93. https://doi.org/10.1002/cam4.1889 . Kastenhuber ER, Lowe SW. Putting p53 in Context. Cell. 2017;170(6):1062–78. https://doi.org/10.1016/j.cell.2017.08.028 . Ganapathy-Kanniappan S, Geschwind JF. Tumor glycolysis as a target for cancer therapy: progress and prospects. Mol Cancer. 2013;12:152. https://doi.org/10.1186/1476-4598-12-152 . Additional Declarations No competing interests reported. Supplementary Files SupplementalFig.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 10 May, 2026 Reviews received at journal 04 May, 2026 Reviewers agreed at journal 30 Apr, 2026 Reviewers agreed at journal 29 Apr, 2026 Reviews received at journal 26 Apr, 2026 Reviewers agreed at journal 26 Apr, 2026 Reviewers invited by journal 26 Apr, 2026 Editor assigned by journal 26 Jan, 2026 Submission checks completed at journal 26 Jan, 2026 First submitted to journal 24 Jan, 2026 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-8687369","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":632285998,"identity":"71792c35-e15e-4ba3-9886-a59a9dcae687","order_by":0,"name":"Ruolan Tan","email":"","orcid":"","institution":"Chongqing City Hospital of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Ruolan","middleName":"","lastName":"Tan","suffix":""},{"id":632285999,"identity":"4923cb4b-dc72-4690-a717-aa61f6d0d096","order_by":1,"name":"Jiajia Ye","email":"","orcid":"","institution":"The First Affiliated Hospital of Army Medical University","correspondingAuthor":false,"prefix":"","firstName":"Jiajia","middleName":"","lastName":"Ye","suffix":""},{"id":632286000,"identity":"62d84636-b42d-4f1d-a349-27c10dd83ca7","order_by":2,"name":"Zhihong Xu","email":"","orcid":"","institution":"The People’s Hospital of Dadukou District","correspondingAuthor":false,"prefix":"","firstName":"Zhihong","middleName":"","lastName":"Xu","suffix":""},{"id":632286001,"identity":"d6cf55c2-5178-4206-a529-23d02e8874c7","order_by":3,"name":"Han Tang","email":"data:image/png;base64,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","orcid":"","institution":"The Affiliated Hospital of Stomatology of Chongqing Medical University","correspondingAuthor":true,"prefix":"","firstName":"Han","middleName":"","lastName":"Tang","suffix":""}],"badges":[],"createdAt":"2026-01-24 14:23:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8687369/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8687369/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108977490,"identity":"55a77340-9f19-469f-8633-f06970abc933","added_by":"auto","created_at":"2026-05-11 11:31:54","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":14165682,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eData analysis process. A.\u003c/strong\u003e Date collection and identification of key prognostic factors. \u003cstrong\u003eB.\u003c/strong\u003e Construction of a prognostic signature based on glycolysis-related genes and evaluation the differences in overall survival and energy metabolism patterns. \u003cstrong\u003eC.\u003c/strong\u003e Comparison of differences in gene expression and mutation. \u003cstrong\u003eD.\u003c/strong\u003e Assessment the therapy response.\u003c/p\u003e","description":"","filename":"Fig.1.png","url":"https://assets-eu.researchsquare.com/files/rs-8687369/v1/2419740c7fa8c87ba6a5fd16.png"},{"id":108839613,"identity":"19aca455-2d30-4cd9-b800-8815a4d56058","added_by":"auto","created_at":"2026-05-09 00:48:59","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":6390198,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDetection of prognostic factors.\u003c/strong\u003e \u003cstrong\u003eA.\u003c/strong\u003e Multivariate Cox regression analysis revealed that glycolysis can be identified as a significant risk factor influencing the prognosis of OSCC patients. \u003cstrong\u003eB.\u003c/strong\u003e A heatmap visually represents the correlations between glycolysis ssGSEA scores and various clinicopathological features, as well as mutations of driver oncogenes. \u003cstrong\u003eC.\u003c/strong\u003e WGCNA was employed to identify glycolysis-related genes.\u003c/p\u003e","description":"","filename":"Fig.2.png","url":"https://assets-eu.researchsquare.com/files/rs-8687369/v1/fe87e7953db43b1444ef2c3c.png"},{"id":108839615,"identity":"0abcdb7c-4c55-4266-a7ca-69ceca7f910c","added_by":"auto","created_at":"2026-05-09 00:48:59","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":12328678,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eConstruction of a glycolysis-related prognostic model and evaluation of survival and metabolic differences. A. \u003c/strong\u003eUtilizing the\u003cstrong\u003e \u003c/strong\u003eLASSO Cox regression algorithm, the most robust prognostic genes were identified. As the tuning parameter (λ) increased, the coefficients of corresponding genes gradually reduced to zero.\u003cstrong\u003e B. \u003c/strong\u003eCross-validation was performed to determine the optimal λ value. \u003cstrong\u003eC.\u003c/strong\u003e Genes involved in the prognostic model. \u003cstrong\u003eD and G.\u003c/strong\u003eScatter plots illustrate the survival status of patients in the High- and Low-risk groups. \u003cstrong\u003eE and H.\u003c/strong\u003e Kaplan-Meier curve analysis demonstrated that patients in the Low-risk group exhibited a better prognosis. \u003cstrong\u003eF and I.\u003c/strong\u003e Significant differences in energy metabolism patterns were observed between High- and Low-risk groups.\u003c/p\u003e","description":"","filename":"Fig.3.png","url":"https://assets-eu.researchsquare.com/files/rs-8687369/v1/02f63461f3289776a1fd7313.png"},{"id":108839617,"identity":"b31d6206-cad8-4852-869b-28b5b8e1a1c9","added_by":"auto","created_at":"2026-05-09 00:48:59","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":11123702,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEnergy metabolic map.\u003c/strong\u003e In the High-risk group, less pyruvate entered mitochondria to participate in the TCA cycle. Meanwhile, less energy was generated through oxidative phosphorylation.\u003c/p\u003e","description":"","filename":"Fig.4.png","url":"https://assets-eu.researchsquare.com/files/rs-8687369/v1/72af5d50ea5b4d5b745e89b3.png"},{"id":108839618,"identity":"73615096-e4b2-45a5-a5f7-d599082ff95c","added_by":"auto","created_at":"2026-05-09 00:48:59","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":72881102,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDifferences in gene expression and mutation between High-risk group and Low-risk group.\u003c/strong\u003e \u003cstrong\u003eA.\u003c/strong\u003e Volcano plot reveals that 264 genes were up-regulated and 1,564 genes down-regulated in the High-risk group. \u003cstrong\u003eB.\u003c/strong\u003eFunctional enrichment analysis highlights the biological processes (BPs) involved in the up- and down-regulated. \u003cstrong\u003eC.\u003c/strong\u003e A comparison of immunocyte infiltration levels between the High-risk and Low-risk groups showed a decreased in most immune cell types in the High-risk group. \u003cstrong\u003eD and E.\u003c/strong\u003e The top 20 mutated genes in both High-risk and Low-risk groups are listed. \u003cstrong\u003eF.\u003c/strong\u003eAmong the intersecting mutated genes, \u003cem\u003eTP53\u003c/em\u003ewas the only one that exhibited a significant difference between the High-risk and -Low groups. \u003cstrong\u003eG.\u003c/strong\u003e A lollipop plot visualizes the different mutation sites of \u003cem\u003eTP53\u003c/em\u003e in the two groups.\u003c/p\u003e","description":"","filename":"Fig.5.png","url":"https://assets-eu.researchsquare.com/files/rs-8687369/v1/f298b638706cf3c3c674d422.png"},{"id":108977104,"identity":"2b7fd37d-07f8-4e3a-b801-3a9ee027bc4d","added_by":"auto","created_at":"2026-05-11 11:30:25","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":33063670,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAssociation between glycolysis-derived signature and therapeutic response. A. \u003c/strong\u003eGSEA predicted that high Z-score is positively correlated with drug resistance. \u003cstrong\u003eB.\u003c/strong\u003eFive chemotherapy drugs that have been applied or studied in OSCC are listed. \u003cstrong\u003eC. \u003c/strong\u003eThe chemotherapeutic sensitivity of these five drugs was estimated and compared between High-risk and Low-risk groups. Statistically significant differences are denoted as * \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05, ** \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01, *** \u003cem\u003ep\u003c/em\u003e\u0026lt; 0.001, **** \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.0001. NS, not statistically significant.\u003c/p\u003e","description":"","filename":"Fig.6.png","url":"https://assets-eu.researchsquare.com/files/rs-8687369/v1/5811398f6b30cdd405c9cfc6.png"},{"id":108839612,"identity":"e088a302-6c1f-4204-8b97-9aa708591284","added_by":"auto","created_at":"2026-05-09 00:48:55","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":259585,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8687369/v1/0341d8ba-78ce-4ef6-a4cd-81b7009ae8ff.pdf"},{"id":109221662,"identity":"fb626fb7-05cb-4204-aea7-ffb93613303b","added_by":"auto","created_at":"2026-05-13 20:51:18","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2350749,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementalFig.docx","url":"https://assets-eu.researchsquare.com/files/rs-8687369/v1/4f578641806bb0acf5c1cee2.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"A Novel Glycolysis Evaluation System for Stratified Management and Individualized Treatment of OSCC Patients","fulltext":[{"header":"Introduction","content":"\u003cp\u003eOral squamous cell carcinoma (OSCC) is a highly recurrent form of cancer arising from the mucosal lining of the oral cavity, characterized by high rates of metastasis, recurrence, and resistance to traditional chemotherapy [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. It is one of the most common malignancies of the head and neck region with ~\u0026thinsp;373,000 new patients diagnosed and ~\u0026thinsp;199,000 deaths reported in 2019 [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Various etiological factors, including smoking, betel nut chewing, immunodeficiency, and alcohol consumption, contribute to mutational changes, which further result in the appearance of OSCC. Conventional therapies of OSCC are surgery, chemotherapy, radiotherapy or a combination of these modalities. Despite advances in cancer diagnosis and treatment, the overall survival of OSCC patients has not improved substantially over the past four decades [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. One-third of OSCC patients eventually develop life-threatening and incurable recurrent disease [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Due to patient heterogeneity, cancer metastasis, neoplasm recurrence, and drug resistance, tumors at similar stages respond very differently to the same treatment. Thus, if proper characteristics can be used to stratify patients for precision and personalized management plans, it will greatly improve the prognosis of OSCC patients [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn 2022, Hanahan proposed fourteen hallmarks in the development of tumors, emphasizing the role of reprogramming energy metabolism in promoting tumor progression [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. The rapid and unbridled proliferation characteristic of tumor growth is an energy- and resource-consuming process, and thus metabolism is significantly altered during neoplastic transformation and tumor progression [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Under aerobic condition, normal cells produce energy primarily through mitochondria, and pyruvate produced by glycolysis entering the mitochondria for the subsequent tricarboxylic acid (TCA) cycle and oxidative phosphorylation (OXPHOS). in contrast, under anaerobic conditions, glycolysis is more advantageous, and pyruvate is primarily converted to lactate with little transfer to the mitochondria. This state is known as the Warburg effect or aerobic glycolysis, first described in the 1920s [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. These alterations cause most cancers to induce unregulated glucose fermentation pathways for energy and to fuel growth, a factor underlying the use of \u003csup\u003e18\u003c/sup\u003eF-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) as an important diagnostic tool for oncologists [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eBased on this characteristic, Abd El-Hafez et al. [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e] calculated a total lesion glycolysis (TLG) of each OSCC patient from PET/CT in a prospective study and observed higher rates of distant metastases and worse prognosis in patients with high TLG. Furthermore, Ryu et al. [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e] retrospectively evaluated the ability of TLG to stratify the likelihood of survival and predict occult metastasis in OSCC, finding that patients with high TLG had inferior outcomes, providing guidance for the formulation of treatment and follow-up strategies. However, the absence of a real quantized gold standard and high price limit the application and promotion of PET/CT-based prognostic assessment. Despite increasing evidence demonstrating that glycolysis can promote the growth, invasion and migration of tumor cells [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], the effect of glycolysis on patient's long-term prognosis and the underlying mechanistic details pertinent to the causes and consequences of such metabolic phenotype remain unclear.\u003c/p\u003e \u003cp\u003eIn this study, we explored the influence of fourteen hallmarks of cancer on the overall survival of OSCC patients. Among them, glycolysis was identified as a risk factor for the prognosis of OSCC. Subsequently, glycolysis-related genes were detected, and a prognosis model was constructed based on these genes. According to the model, patients were separated into two groups, which exhibited different energy metabolic patterns, infiltration levels of immunocytes, gene expression and mutation, response to chemotherapeutic drugs, and prognosis, suggesting a potential approach to stratify and manage patients (the data analysis process is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Moreover, the glycolysis-related model genes have research prospects as new biomarkers for regulating metabolic reprogramming and tumor progression, providing a reference for a new quantitative method of glycolysis.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData acquisition and process\u003c/h2\u003e \u003cp\u003eA total of three OSCC public cohorts with comprehensive gene expression profiles and clinical information were included in this study. Two datasets (GSE41613 and GSE42743) from one study were acquired from Gene Expression Omnibus (GEO) database [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. The raw CEL files were downloaded and integrated into a new cohort using robust multiarray method for data normalization and the Combat algorithm to eliminate batch effects. The combined dataset of 171 OSCC samples was used as a training set. The validation cohort consisted of 303 patients with whole RNA-sequencing data, age, gender, tumor stage, and overall survival information was collected from UCSC Xena (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://xena.ucsc.edu\u003c/span\u003e\u003cspan address=\"http://xena.ucsc.edu\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Somatic mutation data, sorted in Mutation Annotation Format (MAF), were obtained from The Cancer Genome Atlas (TCGA, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://portal.gdc.cancer.gov\u003c/span\u003e\u003cspan address=\"https://portal.gdc.cancer.gov\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Probe IDs were mapped to gene symbols according to the corresponding annotation file, and probes targeting genes in common were summarized by selecting those showing the highest mean of expression between all samples.\u003c/p\u003e \u003cp\u003eFourteen main gene sets of cancer-related hallmarks [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] were retrieved from Molecular Signatures Database (MSigDB, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.gsea-msigdb.org/gsea/msigdb/index.jsp\u003c/span\u003e\u003cspan address=\"https://www.gsea-msigdb.org/gsea/msigdb/index.jsp\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and previously published literatures [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eStatistical methods\u003c/h3\u003e\n\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eConstruction and Evaluation of an Energy Metabolism-Related Prognosis Model\u003c/h2\u003e \u003cp\u003eThe single-sample gene set enrichment analysis (ssGSEA) algorithm was used by R package \u0026ldquo;gsva\u0026rdquo; to calculate the levels of cancer-related hallmarks in each training sample. Multivariate Cox regression, realized by R package \u0026ldquo;survminer\u0026rdquo;, was performed to evaluate the prognosis value of the signatures. To further explore the key hallmark (glycolysis) associated genes, we carried out Weighted correlation network analysis (WGCNA) using the R package \u0026ldquo;WGCNA\u0026rdquo;.\u003c/p\u003e \u003cp\u003eNext, the least absolute shrinkage and selection operator (LASSO) Cox regression analysis was applied using the R package \u0026ldquo;glmnet\u0026rdquo; to narrow down the variates, screen the most robust candidates, and build a glycolysis-related prognostic model. Additionally, Kaplan-Meier curve analysis was used to determine the value of the model for prognostic assessment.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eDifferential, Functional Enrichment, Immune Infiltration and Mutation Analyses\u003c/h3\u003e\n\u003cp\u003eThe R package \u0026ldquo;DESeq2\u0026rdquo; was used to identify differentially expressed genes (DEGs). Genes with a log\u003csub\u003e2\u003c/sub\u003e fold change (log\u003csub\u003e2\u003c/sub\u003e FC)\u0026thinsp;\u0026gt;\u0026thinsp;1 or \u0026lt; -1 and an adjust \u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were identified as DEGs. Functional enrichment analysis was implemented in Cytoscape software 3.7.2. For mutation data, the R package \u0026ldquo;maftools\u0026rdquo; was used to investigate mutation sites and frequencies.\u003c/p\u003e \u003cp\u003eTo quantify the relative proportions of infiltrating immune cells from the gene expression profiles, the R package \u0026ldquo;CIBERSORT\u0026rdquo; was used. The putative abundance of immune cells was estimated using a reference set with 22 types of immune cell subtypes (LM22) with 1,000 permutations.\u003c/p\u003e\n\u003ch3\u003eGene Set Enrichment Analysis (GSEA) and Assessment of Therapeutic Response\u003c/h3\u003e\n\u003cp\u003eThe gene sets associated with drug resistance were acquired from MSigDB, and GSEA was evaluated using the R package \u0026ldquo;clusterProfiler\u0026rdquo;. Furthermore, the R package \u0026ldquo;oncoPredict\u0026rdquo; was applied to estimate the chemotherapeutic responses of patients.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eGlycolysis Identified as an Important Prognostic Risk Factor for Overall Survival in OSCC\u003c/h2\u003e \u003cp\u003eThe prognostic value of the hallmarks was estimated by multivariate Cox regression analysis. Remarkably, glycolysis was the only significant risk factor for overall survival among various cancer-related hallmarks, with a HR\u0026thinsp;=\u0026thinsp;10.445 and a \u003cem\u003ep-\u003c/em\u003evalue\u0026thinsp;=\u0026thinsp;0.042 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). Subsequently, the heatmap exhibited close correlations between overall survival status and tumor stage with glycolysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eExploration of Glycolysis-Associated Candidate Genes\u003c/h3\u003e\n\u003cp\u003eWGCNA was performed with transcriptome profiling data and glycolysis ssGSEA scores to construct a scale-free co-expression network. With a soft threshold of 5, the genes were separated to 44 modules, and the lightgreen module displayed the strongest correlation with glycolysis, indicating it as the source module of candidate genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). Furthermore, LASSO regression helped to narrow down the variates, and twelve genes were selected to construct a prognostic model (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). Among the 12 genes, three (\u003cem\u003ePIK3C2B\u003c/em\u003e, \u003cem\u003eMEX3D\u003c/em\u003e, \u003cem\u003eBMP2\u003c/em\u003e) with coefficients\u0026thinsp;\u0026lt;\u0026thinsp;0 played protective roles, and nine (\u003cem\u003eSLC20A1\u003c/em\u003e, \u003cem\u003eTMEM181\u003c/em\u003e, \u003cem\u003eWDR54\u003c/em\u003e, \u003cem\u003eSNAPC1\u003c/em\u003e, \u003cem\u003eS1PR5\u003c/em\u003e, \u003cem\u003eSTC2\u003c/em\u003e, \u003cem\u003eANXA5\u003c/em\u003e, \u003cem\u003eTHBS1\u003c/em\u003e, \u003cem\u003eKLF7\u003c/em\u003e) were identified as risk factors (coefficients\u0026thinsp;\u0026gt;\u0026thinsp;0, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). According to the formula, risk score = \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\sum\\:_{i=1}^{n}{Coefficietn\\:RNA}_{i}\\:\\times\\:\\:Expression\\:{RNA}_{i}\\)\u003c/span\u003e\u003c/span\u003e, the risk score of each patient was calculated and then normalized to Z-score. Patients with a risk Z-score \u0026gt; 0 were identified as High-risk group and others were classified into Low-risk group.\u003c/p\u003e \u003cp\u003eThe scatter diagram revealed the survival time and status of patients in different groups. The heatmap further exhibited the expression levels of the candidate genes in training and validation cohorts. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG, most genes were significantly different between High- and Low-risk groups, except for \u003cem\u003ePIK3C2B\u003c/em\u003e in the validation cohort. Kaplan-Meier curve analysis demonstrated that the Low-risk group had a better outcome compared with the High-risk group (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH).\u003c/p\u003e \u003cp\u003eSince this prognostic model was constructed based on glycolysis-related candidate genes, we further focused on the differences in energy metabolic patterns between the High- and Low-risk groups. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eI, the High-risk group may primarily generate energy through glycolysis, while the Low-risk group may tend to obtain energy through the TCA cycle and OXPHOS.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eDepicting of Energy Metabolic map with Differentially Expressed Genes\u003c/h2\u003e \u003cp\u003eTo further elucidate the variational trend of related genes during glycolysis, the TCA cycle and OXPHOS, we drew an energy metabolic map. Genes with the same expression tendency in both GEO and TCGA cohorts and significant differences between the High-risk and Low-risk groups in the TCGA cohort were selected and exhibited in the map (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). As shown in the map, genes regulating the conversion of pyruvate to acetyl-COA were down-regulated in the High-risk group, suggesting that less pyruvate produced by glycolysis was converted to acetyl-COA and entered the TCA cycle. Similarly, the relevant regulatory genes in each OXPHOS complex were down-regulated, indicating that less energy was generated through OXPHOS. Taken together, patients in the High-risk group may be less dependent on the TCA cycle and OXPHOS, further confirming our previous hypothesis about energy metabolic patterns in the High- and Low-risk groups.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eComprehensive Analyses of Genomic Alterations between the High-risk group and Low-risk group\u003c/h2\u003e \u003cp\u003eBased on the previous analysis, OSCC patients were preliminarily divided into High-risk and Low-risk groups, and the potential differences in energy metabolic patterns and prognostic survival were observed between the two groups. Next, we analyzed gene expression and mutation differences between the two groups to further explore the underlying mechanisms. There were 264 genes up-regulated and 1,564 genes down-regulated in the High-risk group (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). Functional enrichment analysis of the increased and decreased genes, respectively, revealed that the up-regulated genes were mainly enriched in angiogenesis, epithelial migration, and metallopeptidase activity etc., which were considered to induce or enhance tumor growth, invasion, and metastasis [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. In contrast, the down-regulated genes mainly involved in immune response not conducive to tumor progression, such as epithelial development and keratinization (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB).\u003c/p\u003e \u003cp\u003eSubsequently, CIBERSORT was used to investigate the different infiltration levels of immunocytes between the two groups, and immunocytes with more than 60% of samples scoring zero were eliminated. Most of the immune cells were significantly infiltrated at higher levels in the Low-risk group, including plasma cells, CD8\u003csup\u003e+\u003c/sup\u003e T cells, follicular helper T cells, regulatory T cells (Tregs), and resting dendritic cells. Two immunocytes, resting memory CD4\u003csup\u003e+\u003c/sup\u003e T cells and M0 macrophage, exhibited higher infiltration levels in High-risk group (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). Additionally, Spearman\u0026rsquo;s correlation analysis revealed that the follicular helper T cells, Tregs and CD8\u003csup\u003e+\u003c/sup\u003e T cells were the top three immune cells negatively correlated with risk Z-score, while M0 macrophage showed the strongest positive correlation with the risk Z-score. CD8\u003csup\u003e+\u003c/sup\u003e T cells are crucial members of cytotoxic T cells that kill tumor cells, and their decreased number may lead to a weakened eliminating effect on tumor cells. Further analysis found that the High-risk group had a higher T cell exclusion potential of the tumor (Supplementary Fig.\u0026nbsp;1), indicating that more T cell exclusion may occur in the High-risk group, leading to a lower T cell infiltration level.\u003c/p\u003e \u003cp\u003eThere is evidence that gene mutations play a crucial role in tumor angiogenesis, invasion and migration [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Therefore, we further focused on the differences in mutations between the High-risk and Low-risk groups. Figure\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD and \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE showed the top 20 mutated genes in the two groups. \u003cem\u003eTP53\u003c/em\u003e, \u003cem\u003eTTN\u003c/em\u003e, \u003cem\u003eFAT1\u003c/em\u003e, \u003cem\u003eCDKN2A\u003c/em\u003e, \u003cem\u003eNOTCH1\u003c/em\u003e ranked in the top 5, with \u003cem\u003eTP53\u003c/em\u003e being the most frequently mutated gene in both groups. Fifteen identical mutated genes among the top 20 genes in both the High-risk and Low-risk groups were selected for further analysis. Higher mutation rates were observed in \u003cem\u003eTP53\u003c/em\u003e in the High-risk group compared with Low-risk group (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eF). The lollipop plot showed the differences in \u003cem\u003eTP53\u003c/em\u003e mutation sites between the two groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eG). To sum up, the significant differences in angiogenesis, cell migration, immune response and mutated genes may be related to differences in energy metabolism and survival between the High-risk group and Low-risk group.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eEnergy Metabolic Pattern Associated with Therapeutic Response\u003c/h2\u003e \u003cp\u003eManagement of OSCC often requires multimodality treatments, including surgery, radiation, and/or chemotherapy. Previous studies have expounded that mutations in \u003cem\u003eTP53\u003c/em\u003e are significantly associated with poor survival and tumor resistance to radiotherapy and chemotherapy in OSCC patients [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. Furthermore, glycolysis and glycolytic enzymes have been regarded as pivotal factors promoting a drug-resistant phenotype [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Therefore, we speculated that the differences in \u003cem\u003eTP53\u003c/em\u003e mutation and energy metabolic patterns between the High-risk and Low-risk groups might cause different chemotherapeutic sensitivity to chemotherapy drugs in the two groups. Based on the altered gene sets of different drug treatments retrieved, GSEA predicted that resistance to doxorubicin, gefitinib, cisplatin and imatinib was significantly up-regulated, and resistance to gemcitabine, trabectedin and tamoxifen was down-regulated in the High-risk group, indicating a close correlation between drug resistance and energy metabolism pattern (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). Furthermore, we focused on five drugs that have been used and/or studied in OSCC chemotherapy (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). By estimating the IC50 value of each patient, we compared the chemotherapeutic sensitivity between the two groups. Patients in the High-risk group had higher IC50 values for doxorubicin and gefitinib and lower IC50 values for tamoxifen, consistent with the GSEA results. However, drug sensitivity to cisplatin exhibited no difference, and opposite results were perceived for two types of gemcitabine (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC). These results denoted that energy metabolic patterns have the potential to affect patients\u0026rsquo; chemotherapeutic sensitivity to drugs, which may further result in the different prognosis.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eExpanding incidence of OSCC malignancies and late-stage presentation have made it global healthcare issues. Traditionally, the management of OSCC begins with initial staging using the TNM system, which, however, is insufficient for predicting disease prognosis [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. This underscores the critical need for precise identification and stratification of these malignant lesions to maximize patient outcomes.\u003c/p\u003e \u003cp\u003eHerein, we first validated the prognostic significance of glycolysis in OSCC (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). Cancer cells often reprogram their glucose metabolic pathways to adapt to environmental challenges and support survival, proliferation, and metastasis [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Previous studies have suggested that glycolysis is the major glucose metabolic pathway in OSCC cell lines, with inhibited glycolysis activating intrinsic apoptosis and accelerated glycolysis promoting in vivo tumor development [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Thus, it is not surprising that the glycolysis adversely affects the prognosis of OSCC patients, as observed in our study. After establishing glycolysis as a prognosis risk factor for OSCC, we further investigated the genes associated glycolysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). Based on these genes (\u003cem\u003ePIK3C2B\u003c/em\u003e, \u003cem\u003eMEX3D\u003c/em\u003e, \u003cem\u003eBMP2\u003c/em\u003e, \u003cem\u003eSLC20A1\u003c/em\u003e, \u003cem\u003eTMEM181\u003c/em\u003e, \u003cem\u003eWDR54\u003c/em\u003e, \u003cem\u003eSNAPC1\u003c/em\u003e, \u003cem\u003eS1PR5\u003c/em\u003e, \u003cem\u003eSTC2\u003c/em\u003e, \u003cem\u003eANXA5\u003c/em\u003e, \u003cem\u003eTHBS1\u003c/em\u003e, \u003cem\u003eKLF7\u003c/em\u003e), we developed a glycolysis-related prognostic model (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). Using this model, patients were stratified into High- and Low-risk groups, and distinct outcomes and energy metabolic patterns (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE, \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF, \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH, and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eI). Specifically, the High-risk groups had a poorer prognosis and relied primarily on glycolysis for energy metabolism. Previous studies have attempted to assess tumor glycolysis levels by PET/CT to predict patient prognosis [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. However, the complexity and cost of this method limits its application. The calculation formula presented in this study offers a novel tool that is more convenient, more efficient, more affordable and more accessible for assessing glycolysis levels and predicting OSCC patient outcomes.\u003c/p\u003e \u003cp\u003eIt is well recognized that in tumor cells prefer glycolysis as their primary energy source, even in the presence of sufficient oxygen. This results in increased lactate production and reduced pyruvate entry into mitochondria, leading to decrease mitochondrial energy metabolism [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Our study also observed this phenomenon, with down-regulated expression of genes catalyzing pyruvate oxidation to acetyl-CoA in the High-risk group, reducing acetyl-CoA production (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). As a crucial driver of the TCA cycle, acetyl-CoA is essential for sustaining TCA cycle activity. Reduced acetyl-CoA formation directly impacts subsequent mitochondrial energy metabolism [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. With weakened TCA cycle activity, the amount of succinate entering OXPHOS decreased, and the expressions of OXPHOS complex I-V genes were down-regulated, further limiting energy production through OXPHOS. Despite generating less energy than OXPHOS, glycolysis remains the preferred way for tumor cell proliferation and growth due to its efficient glucose uptake and productivity [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. These findings partially explain why the High-risk group, with high glycolysis and low OXPHOS as their primary energy metabolic pattern, had a worse prognosis than the Low-risk group.\u003c/p\u003e \u003cp\u003eFurthermore, we found that the genes with higher expression in the High-risk group were enriched in biological processes conductive to tumor development, such as cell adhesion mediated by integrins, epithelial cell migration, metallopeptidase and metalloendopeptidase activity, and angiogenesis. Integrins are involved in nearly every step of cancer progression, supporting tumor invasion, acquisition of tumor stem cell properties, and drug resistance [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. Moreover, extensive research has identified matrix metalloproteinases (MMPs) as the most prominent proteinases family associated with tumorigenesis. MMP activity can induce or enhance tumor survival, invasion and metastasis [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Consequently, the elevated expression of these associated genes suggests that tumors in patients with a High-risk score may exhibit faster growth and more aggressive behavior compared to the Low-risk group. On the other hand, the down-regulated genes are primarily involved in epithelial cell differentiation, keratinocyte differentiation, arachidonic acid monooxygenase activity, and humoral immune response. Dyskeratosis is a prognostic histopathological feature that impacts the survival outcomes of patients with OSCC. Recent researchers have observed that OSCC patients with minimal or absent keratinization tend to experience more recurrences, lower 5-year disease-free and disease-specific survival rates, and significantly higher lymph node metastasis compared to those with good or high keratinization [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. On the other hand, host immunity represents the human body\u0026rsquo;s most crucial defensive mechanism. Immune cells within the tumor microenvironment (TME) can either eliminate cancerous cells or promote tumor growth by secreting anti-tumor cytokines, growth factors, or pro-inflammatory factors [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. The TME\u0026rsquo;s hostile environment for anti-tumor immune cells is partly driven by a cascade of biochemical reactions leading to local acidity [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. Preliminary evidence indicates that low pH can facilitate immune escape by both attenuating antitumor effectors and promoting the recruitment and activity of immunosuppressive pro-tumor immune cells [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. Given our previous observations, we speculated that since the High-risk group relies primarily on glycolysis as their metabolic mode, more pyruvate is converted into lactate, altering the local pH and impacting immune cell function. Collectively, these findings further affirm that the High-risk group possesses more advantages in the tumor formation and development process.\u003c/p\u003e \u003cp\u003eGiven the previous results suggesting potential immune differences between the High-risk and -Low groups, we focused on immune cell infiltration levels. A significant decrease in CD8\u003csup\u003e+\u003c/sup\u003e T cells was observed in High-risk group, with a negative correlation between CD8\u003csup\u003e+\u003c/sup\u003e T cells and EMPS Z-score (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). As a key member of tumor-infiltration T lymphocytes, CD8\u003csup\u003e+\u003c/sup\u003e T cells selectively recognize and eliminate cancer cells by interacting with the cognate T cell receptor (TCR) and the HLA/antigen complex, while also secreting anti-tumor cytokines such as interferon (IFN)-γ, tumor necrosis factor (TNF)-α, interleukin (IL)-17, and IL-2 [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Withal, Shimizu et al. [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e] enunciated that previously untreated OSCC patients with high tumor-infiltration CD8\u003csup\u003e+\u003c/sup\u003e T cells had significantly better outcomes, emphasizing the strong correlation between CD8\u003csup\u003e+\u003c/sup\u003e T cells and the prognosis of OSCC patients. Therefore, the lower infiltration level of CD8\u003csup\u003e+\u003c/sup\u003e T cells may contribute to the poorer prognosis in High-risk group. Further analysis revealed that the decrease of CD8\u003csup\u003e+\u003c/sup\u003e T cells may be related to an increase in T cell exclusion potential (Supplementary Fig.\u0026nbsp;1).\u003c/p\u003e \u003cp\u003eBy comparing the mutation differences, we noted that \u003cem\u003eTP53\u003c/em\u003e mutation were significantly higher in the High-risk group than in the Low-risk group (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eF). \u003cem\u003eTP53\u003c/em\u003e mutations are common in human tumors and are linked to poor patient prognosis in many cancers. The wild-type p53 encoded by \u003cem\u003eTP53\u003c/em\u003e is considered an effective tumor suppressor that promotes cell cycle arrest and apoptosis, thereby inhibiting cell growth and oncogenic transformation [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. However, mutations in \u003cem\u003eTP53\u003c/em\u003e lead to the loss p53\u0026rsquo;s tumor suppressive activities, allowing oncogene-expressing cells to proliferate unabated, directly contributing to cancer development [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Consistent with our observations, the High-risk group with a poorer prognosis showed a higher frequency of \u003cem\u003eTP53\u003c/em\u003e mutations, which may ultimately accelerate tumor progression. Overall, the combined effects of differences in energy metabolism pattern, gene expression, immune infiltration and mutation may explain the distinct prognosis between the two groups.\u003c/p\u003e \u003cp\u003eChemotherapy serves as a crucial adjuvant in the treatment of tumors. Research has demonstrated that a heightened glycolytic rate in tumor cells contributes to an increased resistance to chemotherapeutic agents [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Furthermore, mutations in the \u003cem\u003eTP53\u003c/em\u003e gene can lead to enhanced resistance to chemotherapy drugs [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. Does the present research exhibit a similar trait? We conducted a further analysis of the therapeutic response to chemotherapy drugs in both the High-risk and Low groups. We selected five drugs that have been applied and/or investigated in the chemotherapy of OSCC and found that the High-risk group did not display heightened resistance to all chemotherapy drugs (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA and \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC). For cisplatin, a commonly prescribed chemotherapeutic agent for OSCC, no significant increase in resistance or decrease in drug sensitivity was observed in the High-risk group. However, for doxorubicin and gefitinib, there may be an increase in resistance. Although these findings do not fully align with our initial hypotheses, they still offer valuable insights for the selection of chemotherapy management strategies for OSCC patients.\u003c/p\u003e \u003cp\u003eBased on glycolysis-related genes, this study constructed a more convenient, more economical and more generalized glycolysis computational model and initially explored its ability to classify patients and predict patient outcomes. We discovered that patients who rely on glycolysis as their primary energy metabolic mode tend to have a poorer prognosis. This may be associated with gene expression levels, mutation status, as well as immune status. Additionally, we conducted a preliminary investigation into the potential causes of glycolysis affecting patients\u0026rsquo; prognosis and aimed to propose a new target for intervening in tumor glycolysis. Simultaneously, we endeavor to introduce a novel method for evaluating tumor metabolic pattern and patient stratification, offering a fresh perspective for personalized patient treatment.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the National Natural Science Foundation of China (Grant No.82501120). We would like to thank all the study team members for their support and contribution in any form.\u003cstrong\u003e\u003cbr clear=\"all\"\u003e \u003c/strong\u003e\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 (Grant No.82501120). \u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization, Methodology, Supervision: Zhihong Xu and Han Tang; Data curation, Investigation: Ruolan Tan and Han Tang; Formal analysis, Software, Validation, Visualization: Ruolan Tan and Jiajia Ye; Writing \u0026ndash; original draft: Ruolan Tan and Jiajia Ye; Writing \u0026ndash; review \u0026amp; editing: Zhihong Xu and Han Tang. All authors have read and approved the final manuscript.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eData Availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets generated and analyzed during the current study are available in UCSC Xena (http://xena.ucsc.edu) and Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo/), reference number [GSE41613 and GSE42743].\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eThe authors have no relevant financial or non-financial interests to disclose.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to participate\u003c/strong\u003e N/A.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthical approval\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eN/A.\u003c/p\u003e\n"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eChai AWY, Lim KP, Cheong SC. Translational genomics and recent advances in oral squamous cell carcinoma. Semin Cancer Biol. 2020;61:71\u0026ndash;83. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.semcancer.2019.09.011\u003c/span\u003e\u003cspan address=\"10.1016/j.semcancer.2019.09.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGlobal Burden of Disease Cancer, Incidence C, et al. Mortality, Years of Life Lost, Years Lived With Disability, and Disability-Adjusted. JAMA Oncol. 2022;8(3):420\u0026ndash;44. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1001/jamaoncol.2021.6987\u003c/span\u003e\u003cspan address=\"10.1001/jamaoncol.2021.6987\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Life Years for 29 Cancer Groups From 2010 to 2019: A Systematic Analysis for the Global Burden of Disease Study 2019.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZini A, Czerninski R, Sgan-Cohen HD. Oral cancer over four decades: epidemiology, trends, histology, and survival by anatomical sites. J Oral Pathol Med. 2010;39(4):299\u0026ndash;305. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1600-0714.2009.00845.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1600-0714.2009.00845.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLing Z, Cheng B, Tao X. Epithelial-to-mesenchymal transition in oral squamous cell carcinoma: Challenges and opportunities. Int J Cancer. 2021;148(7):1548\u0026ndash;61. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/ijc.33352\u003c/span\u003e\u003cspan address=\"10.1002/ijc.33352\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRyu IS, et al. Prognostic significance of preoperative metabolic tumour volume and total lesion glycolysis measured by (18)F-FDG PET/CT in squamous cell carcinoma of the oral cavity. Eur J Nucl Med Mol Imaging. 2014;41(3):452\u0026ndash;61. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00259-013-2571-z\u003c/span\u003e\u003cspan address=\"10.1007/s00259-013-2571-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. .doi:.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHanahan D. Hallmarks of Cancer: New Dimensions. Cancer Discov. 2022;12(1):31\u0026ndash;46. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1158/2159-8290.CD-21-1059\u003c/span\u003e\u003cspan address=\"10.1158/2159-8290.CD-21-1059\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePoff A, et al. Targeting the Warburg effect for cancer treatment: Ketogenic diets for management of glioma. Semin Cancer Biol. 2019;56:135\u0026ndash;48. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.semcancer.2017.12.011\u003c/span\u003e\u003cspan address=\"10.1016/j.semcancer.2017.12.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWarburg O, Wind F, Negelein E. The Metabolism of Tumors in the Body. J Gen Physiol. 1927;8(6):519\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1085/jgp.8.6.519\u003c/span\u003e\u003cspan address=\"10.1085/jgp.8.6.519\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbd El-Hafez YG, et al. Total lesion glycolysis: a possible new prognostic parameter in oral cavity squamous cell carcinoma. Oral Oncol. 2013;49(3):261\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.oraloncology.2012.09.005\u003c/span\u003e\u003cspan address=\"10.1016/j.oraloncology.2012.09.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGong X, Tang H, Yang K. PER1 suppresses glycolysis and cell proliferation in oral squamous cell carcinoma via the PER1/RACK1/PI3K signaling complex. Cell Death Dis. 2021;12(3):276. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41419-021-03563-5\u003c/span\u003e\u003cspan address=\"10.1038/s41419-021-03563-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi M, et al. Tanshinone IIA inhibits oral squamous cell carcinoma via reducing Akt-c-Myc signaling-mediated aerobic glycolysis. Cell Death Dis. 2020;11(5):381. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41419-020-2579-9\u003c/span\u003e\u003cspan address=\"10.1038/s41419-020-2579-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLohavanichbutr P, et al. A 13-gene signature prognostic of HPV-negative OSCC: discovery and external validation. Clin Cancer Res. 2013;19(5):1197\u0026ndash;203. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1158/1078-0432.CCR-12-2647\u003c/span\u003e\u003cspan address=\"10.1158/1078-0432.CCR-12-2647\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShi R, et al. Identification and validation of hypoxia-derived gene signatures to predict clinical outcomes and therapeutic responses in stage I lung adenocarcinoma patients. Theranostics. 2021;11(10):5061\u0026ndash;76. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7150/thno.56202\u003c/span\u003e\u003cspan address=\"10.7150/thno.56202\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiranda A, et al. Cancer stemness, intratumoral heterogeneity, and immune response across cancers. Proc Natl Acad Sci U S A. 2019;116(18):9020\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.1818210116\u003c/span\u003e\u003cspan address=\"10.1073/pnas.1818210116\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKessenbrock K, Plaks V, Werb Z. Matrix metalloproteinases: regulators of the tumor microenvironment. Cell. 2010;141(1):52\u0026ndash;67. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cell.2010.03.015\u003c/span\u003e\u003cspan address=\"10.1016/j.cell.2010.03.015\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSabapathy K, Lane DP. Therapeutic targeting of p53: all mutants are equal, but some mutants are more equal than others. Nat Rev Clin Oncol. 2018;15(1):13\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nrclinonc.2017.151\u003c/span\u003e\u003cspan address=\"10.1038/nrclinonc.2017.151\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLindemann A, et al. Targeting the DNA Damage Response in OSCC with TP53 Mutations. J Dent Res. 2018;97(6):635\u0026ndash;44. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0022034518759068\u003c/span\u003e\u003cspan address=\"10.1177/0022034518759068\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBhattacharya B, Mohd MF, Omar, Soong R. The Warburg effect and drug resistance. Br J Pharmacol. 2016;173(6):970\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/bph.13422\u003c/span\u003e\u003cspan address=\"10.1111/bph.13422\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChakraborty D, et al. A facile gold nanoparticle-based ELISA system for detection of osteopontin in saliva: Towards oral cancer diagnostics. Clin Chim Acta. 2018;477:166\u0026ndash;72. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cca.2017.09.009\u003c/span\u003e\u003cspan address=\"10.1016/j.cca.2017.09.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReinfeld BI, et al. The therapeutic implications of immunosuppressive tumor aerobic glycolysis. Cell Mol Immunol. 2022;19(1):46\u0026ndash;58. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41423-021-00727-3\u003c/span\u003e\u003cspan address=\"10.1038/s41423-021-00727-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeBerardinis RJ, Chandel NS. Fundamentals of cancer metabolism. Sci Adv. 2016;2(5):e1600200. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/sciadv.1600200\u003c/span\u003e\u003cspan address=\"10.1126/sciadv.1600200\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartinez-Reyes I, Chandel NS. Mitochondrial TCA cycle metabolites control physiology and disease. Nat Commun. 2020;11(1):102. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-019-13668-3\u003c/span\u003e\u003cspan address=\"10.1038/s41467-019-13668-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGanapathy-Kanniappan S. Linking tumor glycolysis and immune evasion in cancer: Emerging concepts and therapeutic opportunities. Biochim Biophys Acta Rev Cancer. 2017;1868(1):212\u0026ndash;20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.bbcan.2017.04.002\u003c/span\u003e\u003cspan address=\"10.1016/j.bbcan.2017.04.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHamidi H, Ivaska J. Every step of the way: integrins in cancer progression and metastasis. Nat Rev Cancer. 2018;18(9):533\u0026ndash;48. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41568-018-0038-z\u003c/span\u003e\u003cspan address=\"10.1038/s41568-018-0038-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWolfer S, Elstner S, Schultze-Mosgau S. Degree of Keratinization Is an Independent Prognostic Factor in Oral Squamous Cell Carcinoma. J Oral Maxillofac Surg. 2018;76(2):444\u0026ndash;54. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.joms.2017.06.034\u003c/span\u003e\u003cspan address=\"10.1016/j.joms.2017.06.034\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJiang X, et al. Role of the tumor microenvironment in PD-L1/PD-1-mediated tumor immune escape. Mol Cancer. 2019;18(1):10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12943-018-0928-4\u003c/span\u003e\u003cspan address=\"10.1186/s12943-018-0928-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuber V, et al. Cancer acidity: An ultimate frontier of tumor immune escape and a novel target of immunomodulation. Semin Cancer Biol. 2017;43:74\u0026ndash;89. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.semcancer.2017.03.001\u003c/span\u003e\u003cspan address=\"10.1016/j.semcancer.2017.03.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShimizu S, et al. Tumor-infiltrating CD8(+) T-cell density is an independent prognostic marker for oral squamous cell carcinoma. Cancer Med. 2019;8(1):80\u0026ndash;93. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/cam4.1889\u003c/span\u003e\u003cspan address=\"10.1002/cam4.1889\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKastenhuber ER, Lowe SW. Putting p53 in Context. Cell. 2017;170(6):1062\u0026ndash;78. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cell.2017.08.028\u003c/span\u003e\u003cspan address=\"10.1016/j.cell.2017.08.028\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGanapathy-Kanniappan S, Geschwind JF. Tumor glycolysis as a target for cancer therapy: progress and prospects. Mol Cancer. 2013;12:152. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/1476-4598-12-152\u003c/span\u003e\u003cspan address=\"10.1186/1476-4598-12-152\" targettype=\"DOI\" 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":false,"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":"medical-oncology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"medo","sideBox":"Learn more about [Medical Oncology](https://www.springer.com/journal/12032)","snPcode":"12032","submissionUrl":"https://submission.nature.com/new-submission/12032/3","title":"Medical Oncology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Energy metabolic reprogramming, Oral squamous cell carcinoma, CD8+ T cell, Cancer hallmarks","lastPublishedDoi":"10.21203/rs.3.rs-8687369/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8687369/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eOral squamous cell carcinoma (OSCC) is one of the most common malignant tumors in the head and neck region, characterized by high rates of metastasis and recurrence and resistance to traditional chemotherapy. We included 474 OSCC patients through the Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA). Then we applied weighted correlation network analysis (WGCNA) to identify glycolysis-related genes and constructed a 12-gene prognostic signature related to glycolysis. Subsequently, we used the model formula to divide the patients into two groups, and analyzed the differences in prognosis, energy metabolic patterns, gene expression, gene mutation, immunity, and therapeutic response between the two groups were analyzed to elucidate underlying mechanisms. Our analysis revealed patients with high-risk score exhibited poorer overall survival and a greater reliance on glycolysis for energy production. Additionally, significant differences were observed between the two groups in biological processes such as angiogenesis, epithelial cell migration, metalloendopeptidase activity, keratinization and humoral immune response. The High-risk group displayed lower infiltration levels of CD8\u0026thinsp;+\u0026thinsp;T cells and higher \u003cem\u003eTP53\u003c/em\u003e mutation rates compared with the Low-risk group. Differences in drug resistance and sensitivity were also noted between the two groups. Collectively, Energy metabolic reprogramming influences the prognosis of OSCC patients, likely due to variations in gene expression, gene mutation, and immunocyte infiltration, particularly CD8\u0026thinsp;+\u0026thinsp;T cells.\u003c/p\u003e","manuscriptTitle":"A Novel Glycolysis Evaluation System for Stratified Management and Individualized Treatment of OSCC Patients","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-09 00:48:49","doi":"10.21203/rs.3.rs-8687369/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-10T12:25:18+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-05T01:37:57+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"160377134263341446269576000047265447265","date":"2026-04-30T08:34:08+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"49274934133447397110930156120316503819","date":"2026-04-29T08:01:44+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-27T02:53:08+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"85380691398871573003579344597926259809","date":"2026-04-27T02:36:56+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-27T02:35:13+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-26T05:57:59+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-26T05:57:33+00:00","index":"","fulltext":""},{"type":"submitted","content":"Medical Oncology","date":"2026-01-24T14:06:25+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"medical-oncology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"medo","sideBox":"Learn more about [Medical Oncology](https://www.springer.com/journal/12032)","snPcode":"12032","submissionUrl":"https://submission.nature.com/new-submission/12032/3","title":"Medical Oncology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"37472c34-fbf0-46f7-8c9c-d35e4f3130bd","owner":[],"postedDate":"May 9th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-10T12:25:18+00:00","index":29,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-05T01:37:57+00:00","index":28,"fulltext":""},{"type":"reviewerAgreed","content":"160377134263341446269576000047265447265","date":"2026-04-30T08:34:08+00:00","index":27,"fulltext":""},{"type":"reviewerAgreed","content":"49274934133447397110930156120316503819","date":"2026-04-29T08:01:44+00:00","index":25,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-09T00:48:50+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-09 00:48:49","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8687369","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8687369","identity":"rs-8687369","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}