A dual EMT-ferroptosis gene signature predicts survival and immune infiltration in esophageal squamous cell carcinoma

A dual EMT-ferroptosis gene signature predicts survival and immune infiltration in esophageal squamous cell carcinoma | 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 dual EMT-ferroptosis gene signature predicts survival and immune infiltration in esophageal squamous cell carcinoma Zhidong Wang, Cheng Gong, Ce Chao, Youpu Zhang, Yiongxiang Qian, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6456491/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background : Limited research has been conducted on the interaction between ferroptosis and epithelial-mesenchymal transition (EMT) and their combined effect on esophageal squamous cell carcinoma (ESCC) patient prognosis. The present study aimed to develop a prognostic model based on the impact of ferroptosis and EMT on ESCC prognosis for clinical application. Methods : Gene expression levels and clinical data of ESCC patients were obtained from the GSE53625 dataset in the gene expression omnibus (GEO) database, and the data from the cancer genome atlas (TCGA) were obtained as a validation set. By combining the results of cox regression analysis and least absolute shrinkage and selection operator regression (LASSO) analysis, we selected nine genes associated with prognosis, which were then used to construct a prognostic model. Immune cell infiltration was evaluated using CIBERSORT and single-sample Gene Set Enrichment Analysis methods. Results : Nine key genes were screened to construct ferroptosis and EMT integrated score (FEIS). Compared to the low-FEIS group, the high-FEIS group demonstrated shorter overall survival period. The immune infiltration analysis showed an increase in immune cell infiltration and elevated expression levels of immune checkpoint molecules in the high-FEIS group. A nomogram was constructed to accurately predict patient prognosis. Conclusion : Our study introduced a novel prognostic tool that integrates ferroptosis -and EMT-related biomarker, and offered valuable insights for developing personalized treatment strategies for ESCC patients. ferroptosis epithelial-mesenchymal transition GEO TCGA bioinformatics analysis esophageal squamous cell carcinoma Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Esophageal cancer (EC) is the seventh most common cancer and the sixth leading cause of cancer-related deaths worldwide[ 1 ]. Esophageal squamous cell carcinoma (ESCC) is a prominent subtype of EC, and its adverse prognosis stems from its subtle initial symptoms, propensity for metastasis, resistance to radiotherapy, and tendency for tumor recurrence [ 2 , 3 ]. In recent years, there has been considerable progress in advancing multidisciplinary and surgical therapies for ESCC. However, despite these advancements, the median survival rate remains low for ESCC patients [ 3 ]. Hence, it is essential to search for new biomarkers to promote personalized treatment and improve survival rate and quality of life of ESCC patients. Ferroptosis is a specific cell death program associated with the accumulation of iron ions and uncontrolled oxidative reactions within a cell [ 4 ]. There has been an exponential growth in research on ferroptosis, with numerous studies highlighting its remarkable role in cancer advancement and treatment outcomes [ 5 , 6 ]. Despite extensive efforts to develop novel therapeutic strategies, ESCC treatment presents several complex challenges, including increased chemotherapy drug resistance and reduced sensitivity of ESCC cells to radiotherapy. A study showed that STC2, as a promising therapeutic target, addressed ESCC radioresistance by inactivating PRMT5 to prevent DNA damage repair and inhibition of ferroptosis [ 7 ]. Gong et al. found that targeting metabolic checkpoints could be used as an effective strategy for ESCC therapy by suppressing tumor ferroptosis [ 8 ]. Additionally, another study showed that treatment with 5-aminolevulinic acid suppressed the GPX4 and HMOX1 overexpression, thereby facilitating ferroptosis in ESCC cells [ 9 ]. Hence, focusing on ferroptosis genes and their related pathways could advance research on suppressing tumor growth. Understanding ferroptosis status in various ESCC patients is therefore crucial to improve the effectiveness of personalized precision treatments. Epithelial-mesenchymal transition (EMT) is a pivotal cellular program essential for embryogenesis, wound healing, and malignancy progression [ 10 , 11 ]. In neoplasms, EMT endows cancer cells with high tumor-initiating and metastatic capabilities and also influences the growth of cancer stem cells, tumor immune evasion, and resistance to drugs [ 12 , 13 ]. A previous study showed that EMT and cancer cell stemness were associated with chemotherapy resistance of EC, and addressing both these properties could serve as an improved approach to combat drug resistance in EC [ 14 ]. Research on EMT-like modifications in tumor cells indicates enhanced aggressiveness, with a strong association between EMT transcriptome characteristics and unfavorable prognoses across various cancer patient cohorts [ 15 ]. Considering these characteristics, the complex biological mechanism of EMT is thought to be closely associated with carcinogenesis, and directing interventions toward EMT pathways has emerged as a promising approach for ESCC therapy. The potential association between ferroptosis and EMT, both of which are critical biological processes in cancer cell, is attracting increasing attention. According to previous research, the occurrence of EMT in cancer cells increases their susceptibility to ferroptosis. Bach1, a crucial regulator of glioma invasion, orchestrates various ECM-related functions, and the overexpression of Bach1 reduces ferroptosis in glioma cells [ 16 ]. In melanoma cells, TGF-β1 triggers EMT and simultaneously elevates lipid peroxidation levels, thereby promoting the cellular process of ferroptosis[ 17 ]. Another noteworthy finding is that ferroptosis may increase EMT inhibition. Guan et al. reported that DpdtbA exerted its inhibitory effect on EMT by activating the p53 and PHD2/HIF-1α pathways [ 18 ]. However, Yao et al. revealed that the inclination toward ferroptosis positively influenced EMT occurrence in the epithelial cells of lung adenocarcinoma [ 19 ]. Increasing evidence indicates that the subtle interaction between ferroptosis and EMT plays a substantial role in the biological processes of tumors. Hence, cancer treatment efforts should be directed toward targeting ferroptosis and EMT. In the present research, we are committed to developing a prognostic model that comprehensively considers the impact of ferroptosis and EMT on the prognosis of ESCC. Materials and methods Data collection and acquisition of gene expression data The dataset GSE53625, derived from the GEO database, served as the training set for the analysis. The dataset was accessed via the “GEOquery” package in R, and the “limma” package was employed to process the mRNA expression matrix. This dataset comprised 179 ESCC carcinoma specimens and 179 adjacent healthy tissue specimens. The TCGA database was screened to obtain complete clinical information of the patients as the validation set. After excluding samples with missing data, 94 samples were eligible for the analysis, which included 91 ESCC specimens and 3 adjacent normal tissue specimens. Thus, the gene expression data were obtained from a total of 270 tumor samples and 182 normal tissue samples, accompanied with clinicopathological information (Table 1 ). The gene expression data were subjected to normalization with log2(TPM + 1). We obtained 259 ferroptosis-related genes (FRGs) from the FerrDb database and 1011 protein-coding EMT-related genes (ERGs) from the dbEMT 2.0 database. Table 1 Baseline characteristics of ESCC patients in GEO and TCGA cohorts GEO TCGA (n = 179) (n = 91) Age 60 80 32 Gender Male 146 77 Female 33 14 TNM Stage I-II 87 61 III-IV 92 30 Tobacco use Yes 65 No 114 Alcohol use Yes 106 No 73 Tumor loation Lower 62 Middle 97 Upper 20 Tumor grade Poorly 98 Moderately 49 Well 32 Identification of critical genes Significantly different genes (SDGs) were screened out by comparing tumor samples with normal samples. The cutoff values were set as false discovery rate (FDR) 0.2. Then FRGs and ERGs were then intersected with SDGs, and the overlapping genes were shown with a Venn diagram. Subsequently, univariate Cox regression analysis was conducted to discover potential biomarkers. LASSO regression analysis was employed to simplify the model and ruduce overfitting risk. The “glmnet” package was used to determine the key genes and to generate the corresponding coefficients for these genes. Finally, the ferroptosis and EMT integrated score (FEIS) was determined with the following equation: FEIS = \(\:\sum\:_{i=1}^{n}{coef}_{i}*{geneExp}_{i}\) , where the \(\:{coef}_{i}\) represents the coefficient and \(\:{geneExp}_{i}\) denotes the expression value of each key gene. Based on the median FEIS value, all patients were categorized into high-FEIS group or low-FEIS group. Relationship between key FRGs and ERGs calculated using Pearson correlation coefficients based on gene expression values, and the chord diagram provides an intuitive visualization. Enrichment analysis Gene Set Enrichment Analysis (GSEA) was employed to compare the enrichment of gene sets between the different risk groups. We utilized GSEA software (version 4.3.2) to conduct enrichment analysis, and selected “c2.cp.kegg.v7.5.1.symbols.gmt'” as the gene sets database. The number of permutations was set to 1,000, with the phenotype labels defined as “low FEIS versus high FEIS,” and the enrichment statistic set to “weighted.” To avoid dataset specificity, the same approach using the TCGA dataset. Ensuring the significance and reliability of the results, we applied the following criteria for pathway enrichment: nominal p-value < 0.05 and FDR < 0.25. Immunoinfiltration analysis The single-sample Gene Set Enrichment Analysis (ssGSEA) method, implemented using the R package “GSVA”, was applied to calculate and analyze the enrichment scores of immune-related gene sets in each sample. We performed cluster analysis, dividing the samples into high immune infiltration and low immune infiltration groups based on the clustering results. Additionally, a validated expression matrix of 22 immune cell subtypes (LM22) was utilized to analyze the proportions of immune cells in tumor tissues. Construction and validation of the nomogram We constructed a nomogram to forecast the clinical outcomes for ESCC patients in the TCGA and GEO databases. TNM stage and FEIS were included as the prognostic factors. Receiver operating characteristic (ROC) curves were generated to validate the constructed nomogram. Statistical analysis Statistical analysis was performed using GraphPad (version 9.5.1) and R software (version 4.3.0). Categorical variables were compared using the chi-square test. The levels of gene expression, immune cell infiltration, and immune pathway activation between the different subgroups were analyzed using Wilcoxon test. Visualization was performed with “rms,” “ggbiplot,” “pheatmap,” “ggplot2,” “survival,” and “ggpubr” packages. p < 0.05 was considered statistically significant. Result Biomarker identification and survival analysis As shown in the flowchart (Fig. 1 ), this study included 1208 FRGs and ERGs. Variations in gene expression were exhibited by a total of 714 genes between adjacent normal tissues and malignant tissues (Fig. 2 A, B). LASSO regression analysis indicated nine genes associated with ferroptosis and EMT (Fig. 2 C, D). These included 6 ERGs ( VIM , HOOK1 , MAP4K4 , PLA2G4A , F2RL2 , and LASP1 ) and 3 FRGs ( ACSL3 , ALOXE3 , and ANGPTL7 ). The chord diagram visually represented FRGs in the green module and ERGs in the pink module. The thickness and color of the ribbons indicated gene expression correlation between these categories, suggesting their interaction (Fig. 2 E). Patients were assigned to the high- and low-FEIS groups based on the midpoint FEIS value (Fig. 2 F), with those in the high-FEIS group showing poorer prognosis and shorter survival duration compared to the low-FEIS group (p < 0.001) (Fig. 2 G, H). According to principal component analysis (PCA), the patient composition differed significantly between the two groups (Fig. 2 I). The areas under the ROC curve (AUC) values at 1-, 3-, and 5- year survival period was 0.737, 0.756, and 0.771, respectively (Fig. 2 J). Model validation in the TCGA cohort The TCGA database cohort was used as a validation set. In line with findings from the GEO cohort, the high-FEIS group showed adverse prognosis and increased number of early deaths (p < 0.05; Fig. 2 K-M). The results of PCA showed clear separation of the patients into two distinct subgroups (Fig. 2 N). The AUC values at 1- and 2-year survival period were 0.638 and 0.764, respectively (Fig. 2 O). Subgroup analysis for the prognostic signature As shown in Table 2 , HOOK1 , PLA2G4A , and ALOXE3 exhibited protective effects on ESCC patients (hazard ratio [HR] < 1, p 1, p < 0.01). The gene expression levels were shown in the heatmap for the subgroups within the GEO cohort (Fig. 3 A). The baseline clinical data between the high- and low-FEIS groups were analyzed, revealing that the high-FEIS group had a significantly greater proportion of patients with stage III-IV cancer compared to the low-FEIS group (p < 0.005) (Table 3 ). Next, the FEIS and TNM stage were included in the univariate Cox analysis to assess their effects on survival time. According to the forest plot, both the FEIS and TNM stage were influential indicators for predicting patient outcomes (GEO cohort: FEIS, p < 0.001; TNM stage, p < 0.01; TCGA cohort: FEIS, p = 0.003; TNM stage, p = 0.013; Fig. 3 B, D). In the multivariate Cox regression analysis, the results indicated that the FEIS was an independent prognostic factor for ESCC patients in both GEO (p < 0.001, HR = 2.775; Fig. 3 C) and TCGA cohorts (p < 0.05, HR = 2.631; Fig. 3 E). Table 2 The 9 genes in the LASSO model Gene HR HR 95L HR 95H p value HOOK1 0.761353 0.636666635 0.910458363 0.002808 PLA2G4A 0.802401 0.696918162 0.923849755 0.002203 ALOXE3 0.820369 0.716824994 0.938869542 0.004024 F2RL2 1.273074 1.093504583 1.482131596 0.001857 ANGPTL7 1.331697 1.103461726 1.607139998 0.002823 MAP4K4 1.48965 1.105592942 2.007119505 0.008797 ACSL3 1.642826 1.200264165 2.248570269 0.001936 VIM 1.756214 1.22448229 2.518849046 0.002209 LASP1 1.974352 1.195652903 3.260196908 0.007854 Table 3 Clinicopathological features of ESCC patients in low- and high-FEIS group GEO TCGA Characteristics Low-FEIS High-FEIS P value Low-FEIS High-FEIS P value (N = 90) (N = 89) (N = 46) (N = 45) Age 1.00 0.125 ≤ 60 50 (55.6%) 49 (55.1%) 26 (56.5%) 33 (73.3%) > 60 40 (44.4%) 40 (44.9%) 20 (43.5%) 12 (26.7%) Gender 0.57 0.773 Male 75 (83.3%) 71 (79.8%) 38 (82.6%) 39 (86.7%) Female 15 (16.7%) 18 (20.2%) 8 (17.4%) 6 (13.3%) TNM Stage 0.02 0.027 I-II 52 (57.8%) 35 (39.3%) 36 (78.3%) 25 (55.6%) III-IV 38 (42.2%) 54 (60.7%) 10 (21.7%) 20 (44.4%) Tobacco use 0.16 - No 28 (31.1%) 37 (41.6%) - - Yes 62 (68.9%) 52 (58.4%) - - Alcohol use 0.29 - No 33 (36.7%) 40 (44.9%) - - Yes 57 (63.3%) 49 (55.1%) - - - tumor location 0.74 Lower 33 (36.7%) 29 (32.6%) - - Middle 46 (51.1%) 51 (57.3%) - - Upper 11 (12.2%) 9 (10.1%) - - tumor grade 0.14 - Moderately 55 (61.1%) 43 (48.3%) - - Poorly 19 (21.1%) 30 (33.7%) - - Well 16 (17.8%) 16 (18.0%) - - Analysis of ferroptosis and EMT status AR, PDK4, MEF2C, ACSL3, USP11, NCOA3, PTPN18, TMSB4X, SLC40A1 , and SIRT1 are known suppressors of ferroptosis (SOFs)[ 20 – 29 ]. We compared the expression patterns of these SOFs between the two subgroups to evaluate ferroptosis status. In the GEO cohort, the expression levels of the AR , PDK4 , MEF2C , ACSL3 , USP11 , NCOA3 , PTPN18 , TMSB4X , SLC40A1 , and SIRT1 genes were significantly increased in the high-FEIS group (Fig. 4 A). Similarly, in the TCGA cohort, the high-FEIS group showed a significant increase in the expression levels of the MEF2C , USP11 , PTPN18 , TMSB4X , SLC40A1 , and SIRT1 genes (Fig. 4 C). These findings indicated a potentially suppressive ferroptosis status in the high-FEIS group. Subsequently, we further analyzed the expression levels of EMT markers within the subgroups. In the GEO cohort, the expression levels of the ZEB1 , TWIST1 , VIM , FN1 , ZEB2 , FOXC2 , PTX3 , WT1 , SEMA3E , NKX3-2 , CYP1B1 , NKX6-1 , SCUBE3 , AGTR1 , and BVES genes were significantly upregulated in the high-FEIS group (Fig. 4 B). In the TCGA cohort, the high-FEIS group showed a notable increase in the gene expression levels of the relevant EMT markers, except for SCUBE3 (Fig. 4 D). These results suggest that the high-FEIS group patients had more active expression of EMT-related genes, thus making them more prone to adverse prognostic events such as tumor metastasis. The GSEA analysis results further confirmed that biological pathways related to EMT were abundant in the high-FEIS group in the GEO and TCGA cohort. These pathways included the overexpression of chondroitin sulfate and actin cytoskeleton, and the enrichment of focal junction and EMC (Extracellular Matrix) receptor interaction pathways in the GEO cohort (Fig. 4 E). Similarly, the pathways related to focal junction and EMC receptor interaction were enriched in the high FEIS group in the TCGA cohort (Fig. 4 F). Additionally, the Hedgehog, Notch3 and PI3K/AKT signaling pathways, which could promote the occurrence of EMT, were upregulated in the high-FEIS group in the TCGA cohort[ 30 – 32 ] (Fig. 4 F). Tumor immune microenvironment in the subgroups. According to the results of ssGSEA, cluster analysis was performed to divide the samples into high immune infiltration and low immune infiltration clusters (Fig. 5 A). A chi-square test was conducted to determine whether the high-FEIS group showed elevated or reduced levels of immune cell infiltration. The results indicated a significant prevalence of high immune infiltration levels in the high-FEIS group (p < 0.05; Supplementary Table 1). According to the CIBERSORT analysis, CD8 T cells, activated CD4 memory T cells, M1 macrophages, and M2 macrophages showed relatively higher abundance in the high-FEIS group. The infiltration levels of monocytes, activated dendritic cells, and activated mast cells were higher in the low-FEIS group (Fig. 5 B). Further comparison of the expression levels of immune checkpoint molecules between the subgroups revealed a notable increase in the expression levels of the BTLA , CTLA4 , ICOSLG , PDCD1LG2 , and TNFRSF9 genes in the high-FEIS group (Fig. 5 C-G). Establishment and validation of the prognostic nomogram We selected the FEIS and TNM stage as prognostic factors to construct a nomogram and evaluated its precision and clinical effectiveness in the GEO and TCGA datasets (Fig. 6 A). Based on the time-dependent ROC curves, the AUC values of the diagnostic model were 0.727 (1-year), 0.779 (3-year), and 0.764 (5-year) in the GEO cohort (Fig. 6 B-D). In the TCGA cohort, because few patients survived beyond 3 years, the AUC values were 0.696 and 0.782 at 1 and 2 years of survival (Fig. 6 E, F). These values indicated that the developed nomogram had reasonable sensitivity and specificity in the GEO and TCGA cohort. The calibration curve showed a moderate consistency between the predicted values of this diagnostic model and the actual values in the GEO (Fig. 6 G-I) and TCGA cohorts (Fig. 6 J, K). DCA (decision curve analysis) indicated that the developed nomogram had dequate clinical utility in both GEO cohort (Fig. 6 L-N) and TCGA cohort (Fig. 6 O, P). Discussion This research comprehensively analyzed how ferroptosis and EMT impact the prognosis for ESCC patients and constructed a predictive nomogram model that incorporated biomarkers related to both these processes. Our proposed signature comprised 6 ERGs ( VIM , HOOK1 , MAP4K4 , PLA2G4A , F2RL2 , and LASP1 ) and 3 FRGs ( ACSL3 , ALOXE3 , and ANGPTL7 ). These genes are highly associated with the onset and development of cancers. VIM is crucial in various biological processes and diseases. This gene encodes the vimentin protein that regulates adhesion among cells and participates in migration and invasion processes[33, 34]. Chao et al. demonstrated that the sex-determining region Y-Box 2 regulated the expression of VIM and the activity of signaling pathways associated with EMT, thereby influencing the phenotypic transition and functional characteristics of cells, which subsequently affects the invasion and metastatic capability of ESCC tumor cells [35]. ACSL3 participates in regulating lipid metabolism and cellular energy balance[36]. Magtanong et al. found that ACSL3 could activate the exogenous monounsaturated fatty acids to reduce the sensitivity pertaining to lethal oxidation in plasma membrane lipids, thereby enhancing cellular resistance to ferroptosis [37]. ACSL3 expression is upregulated across various cancer types, and it predominantly facilitates cancer development through diverse mechanisms. Another study indicated that ACSL3 enhanced the growth and metastatic potential of cancer cells [38]. In the present study, we observed that Hook1 was a protective factor for EC patients; this finding was consistent with a previous report that HOOK1 could suppress renal cell carcinoma progression through the TGF-β and TNFSF13B/VEGF-A pathway [39]. MAP4K4 encodes a serine/threonine kinase that plays a crucial role in cellular signal transduction pathways and promotion of ovarian cancer metastasis [40, 41]. PLA2G4A is a calcium-dependent phospholipase A2 family member involved in cell adhesion, migration, regulation of cell cycle progression, and various physiological processes related to tumor initiation and development. A previous study noted that annexin A10 facilitates metastasis of extrahepatic cholangiocarcinoma by promoting EMT through the PLA2G4A/PGE2/STAT3 pathway [42]. The F2RL2 protein, similar to F2RL3, plays several important roles in cell and tumor biology. A previous study showed that F2RL3 enhanced the migration and invasion abilities of tumor cells by activating different signaling pathways, facilitating their passage through the cell matrix and spread to other tissues and organs, thereby promoting tumor metastasis [43]. The gene ALOXE3 encodes the lipoxygenase protein. Lipoxins are the products of the lipoxygenase catalytic reaction, which play critical roles in cellular signaling, inflammation, immune responses, and other physiological processes. An excessive amount of lipoxins may promote the occurrence of ferroptosis [44, 45]. According to a previous study, the absence of ALOXE3 leads to the onset of malignant glioblastoma, while the inhibition of ALOXE3 expression confers resistance to p53-mediated ferroptosis in glioblastoma cells [46]. Angiopoietin-like proteins (ANGPTLs), with a function in chronic inflammation, are one of the key factors in carcinogenesis, and the proinflammatory factor ANGPTL7 regulates cancer progression [47]. Although the role of ANGPTL7 in ferroptosis is unclear, a previous study reported that the homologous protein ANGPTL3 could regulate the activity of ferroptosis mechanisms and the response of tumor cells to chemotherapeutic drugs in epithelial ovarian cancer through the TAZ-ANGPTL4-NOX2 axis [48]. Our present study also assessed the status of ferroptosis and EMT in both high- and low-FEIS groups and observed a notable difference. Patients in the high-FEIS group showed a notably subdued ferroptosis status paired with enhanced EMT activity. This pattern was further corroborated by the results of GSEA in GEO and TCGA, which highlighted the substantial enrichment of pathways associated with tumor metastasis and invasion specifically in the high-FEIS group. In summary, the suppressed ferroptosis and active EMT state in the high-FEIS group are important potential factors that contribute to the poor prognosis of patients. We used ssGSEA and CIBERSORT for assessing immune cell types and quantities in ESCC tissues, evaluating immune pathway activity levels to enhance our comprehension of the immune microenvironment in ESCC. In line with prior studies, our study delineated two distinct subtypes of immune cell infiltration in the context of ssGSEA results[49]. The high-FEIS cohort had greater immune cell infiltration than the low-FEIS group, thereby providing a basis for the personalized treatment of different subgroups of patients. Tumor-associated macrophages are known to support tumor growth and metastasis. The ssGSEA analysis showed that M2 macrophages were upregulated in the high-FEIS group, which was strongly linked to tumor advancement and invasion [50]. Chen et al. found that more M2 macrophages could migrate into the tumor microenvironment following FOXO1 induction. By inhibiting antitumor immune responses and promoting angiogenesis, these M2 macrophages further supported the survival and spread of tumor cells, leading to a poor patient prognosis [51]. Immune checkpoint molecules can induce immunosuppressive effects in tumors, thus hindering immune responses against them. In the present study, the expression levels of BTLA , CTLA4 , ICOSLG , PDCD1LG2 , and TNFRSF9 were upregulated in the high-FEIS group; This finding further validated the role of the biomarkers included in our nomogram model for risk stratification and prognosis assessment. Previous studies have confirmed that the models based on ferroptosis- or EMT-related genes play important roles in assessing ESCC patient prognosis [52, 53] However, considering the heterogeneity in gene expression and characteristics among different tumor cells and regions in ESCC, predictive methods that rely solely on ferroptosis or EMT-related markers have limitations and may lack high specificity and sensitivity. Compared to the above studies, our study integrated both ferroptosis- and EMT-related genes into a single prognostic model. This combined approach enhanced predictive accuracy compared to models that focused on either pathway alone. Additionally, our model provides the ability to distinguish between different states, including ferroptosis, EMT, and immune suppression, offering a more comprehensive understanding of the tumor microenvironment in ESCC. This contributes valuable insights for personalized treatment strategies, an area that previous models did not address comprehensively. Furthermore, we have included a discussion on the relationship between ACSL3 and VIM expression and ESCC progression, providing a more comprehensive approach to understanding tumor behavior. By comparing our results with these previous studies, the integrated model offers improved predictive power and broader clinical relevance, especially in guiding personalized treatment for ESCC patients. There are several limitations of this study. First, the data used were obtained from public databases, and the included samples may lack broad representativeness, which limitted the generalizability of the research. Second, the current study is limited by its retrospective nature, which may introduce biases related to sample selection and data collection. In conclusion, the present study identified a novel prognostic signature based on EMT and ferroptosis for predicting the overall survival of ESCC patients. Additionally, the developed model reflects the immune status of ESCC patients and offers valuable insights for developing personalized treatment strategies for them. Declarations Data availability statement TCGA: https://portal.gdc.cancer.gov/ GEO database: https://www.ncbi.nlm.nih.gov/geo/ FerrDb website: http://www.zhounan.org/ferrdb dbEMT2.0: http://dbemt.bioinfo-minzhao.org/index.html Author Contribution Zhidong Wang: Drafting the manuscript Cheng Gong: Methodology, Software, Ce Chao: Resources, Software Youpu Zhang: Methodology Yongxiang Qian: Investigation, Methodology Min Wang: Software, Funding acquisition Bin Wang: review and editing of the manuscript Yang Liu: review and editing of the manuscript Funding This work was supported by Top Talent of Changzhou "14th Five-Year Plan" High-level Health Personnel Training Project (KY20221388); Major projects of the Changzhou Health Commission (ZD202205); Changzhou Science and Technology Support Plan (Social Development) (CE20205039). Declaration of Interest statement None Ethics and consent statement This study received approval from the Clinical Research Ethics Committee of the Third Affiliated Hospital of Soochow University, and informed consent was obtained from all participants. Ethics approval number: 2023(Teaching)CL045-02. Clinical trial number: not applicable. Consent for publication Not applicable References Sung, H., et al., Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin, 2021. 71 (3): p. 209-249. Reichenbach, Z.W., et al., Clinical and translational advances in esophageal squamous cell carcinoma. Adv Cancer Res, 2019. 144 : p. 95-135. Morgan, E., et al., The Global Landscape of Esophageal Squamous Cell Carcinoma and Esophageal Adenocarcinoma Incidence and Mortality in 2020 and Projections to 2040: New Estimates From GLOBOCAN 2020. Gastroenterology, 2022. 163 (3): p. 649-658.e2. Jiang, X., B.R. Stockwell, and M. Conrad, Ferroptosis: mechanisms, biology and role in disease. Nat Rev Mol Cell Biol, 2021. 22 (4): p. 266-282. Zhao, L., et al., Ferroptosis in cancer and cancer immunotherapy. Cancer Commun (Lond), 2022. 42 (2): p. 88-116. Hao, M., et al., Ferroptosis regulation by methylation in cancer. Biochim Biophys Acta Rev Cancer, 2023. 1878 (6): p. 188972. Jiang, K., et al., STC2 activates PRMT5 to induce radioresistance through DNA damage repair and ferroptosis pathways in esophageal squamous cell carcinoma. Redox Biol, 2023. 60 : p. 102626. Lv, M., et al., CDK7-YAP-LDHD axis promotes D-lactate elimination and ferroptosis defense to support cancer stem cell-like properties. Signal Transduct Target Ther, 2023. 8 (1): p. 302. Shishido, Y., et al., Antitumor Effect of 5-Aminolevulinic Acid Through Ferroptosis in Esophageal Squamous Cell Carcinoma. Ann Surg Oncol, 2021. 28 (7): p. 3996-4006. Dongre, A. and R.A. Weinberg, New insights into the mechanisms of epithelial-mesenchymal transition and implications for cancer. Nat Rev Mol Cell Biol, 2019. 20 (2): p. 69-84. Debnath, P., et al., Epithelial-mesenchymal transition and its transcription factors. Biosci Rep, 2022. 42 (1). Mittal, V., Epithelial Mesenchymal Transition in Tumor Metastasis. Annu Rev Pathol, 2018. 13 : p. 395-412. Mortezaee, K., J. Majidpoor, and E. Kharazinejad, Epithelial-mesenchymal transition in cancer stemness and heterogeneity: updated. Med Oncol, 2022. 39 (12): p. 193. Liu, X., et al., EMT and Cancer Cell Stemness Associated With Chemotherapeutic Resistance in Esophageal Cancer. Front Oncol, 2021. 11 : p. 672222. Zhao, L., et al., Integrative network biology analysis identifies miR-508-3p as the determinant for the mesenchymal identity and a strong prognostic biomarker of ovarian cancer. Oncogene, 2019. 38 (13): p. 2305-2319. Cong, Z., et al., BTB domain and CNC homolog 1 promotes glioma invasion mainly through regulating extracellular matrix and increases ferroptosis sensitivity. Biochim Biophys Acta Mol Basis Dis, 2022. 1868 (12): p. 166554. Liu, L., et al., Ferroptosis: Mechanism and connections with cutaneous diseases. Front Cell Dev Biol, 2022. 10 : p. 1079548. Guan, D., et al., The DpdtbA induced EMT inhibition in gastric cancer cell lines was through ferritinophagy-mediated activation of p53 and PHD2/hif-1α pathway. J Inorg Biochem, 2021. 218 : p. 111413. Yao, J., et al., Single-Cell RNA-Seq Reveals the Promoting Role of Ferroptosis Tendency During Lung Adenocarcinoma EMT Progression. Front Cell Dev Biol, 2021. 9 : p. 822315. Zhang, W., et al., Resveratrol inhibits ferroptosis and decelerates heart failure progression via Sirt1/p53 pathway activation. J Cell Mol Med, 2023. 27 (20): p. 3075-3089. Bao, Z., et al., MEF2C silencing downregulates NF2 and E-cadherin and enhances Erastin-induced ferroptosis in meningioma. Neuro Oncol, 2021. 23 (12): p. 2014-2027. Klasson, T.D., et al., ACSL3 regulates lipid droplet biogenesis and ferroptosis sensitivity in clear cell renal cell carcinoma. Cancer Metab, 2022. 10 (1): p. 14. Qin, Z., et al., Design and synthesis of isothiocyanate-containing hybrid androgen receptor (AR) antagonist to downregulate AR and induce ferroptosis in GSH-Deficient prostate cancer cells. Chem Biol Drug Des, 2021. 97 (5): p. 1059-1078. Song, X., et al., PDK4 dictates metabolic resistance to ferroptosis by suppressing pyruvate oxidation and fatty acid synthesis. Cell Rep, 2021. 34 (8): p. 108767. Zhu, J., et al., The deubiquitinase USP11 ameliorates intervertebral disc degeneration by regulating oxidative stress-induced ferroptosis via deubiquitinating and stabilizing Sirt3. Redox Biol, 2023. 62 : p. 102707. Qiao, J., et al., NR5A2 synergizes with NCOA3 to induce breast cancer resistance to BET inhibitor by upregulating NRF2 to attenuate ferroptosis. Biochem Biophys Res Commun, 2020. 530 (2): p. 402-409. Hao, L., et al., SLC40A1 Mediates Ferroptosis and Cognitive Dysfunction in Type 1 Diabetes. Neuroscience, 2021. 463 : p. 216-226. Wang, H., et al., Silencing of PTPN18 Induced Ferroptosis in Endometrial Cancer Cells Through p-P38-Mediated GPX4/xCT Down-Regulation. Cancer Manag Res, 2021. 13 : p. 1757-1765. Wang, Z., et al., Protein and metabolic profiles of tyrosine kinase inhibitors co-resistant liver cancer cells. Front Pharmacol, 2024. 15 : p. 1394241. Li, M., et al., Curcumin inhibits the invasion and metastasis of triple negative breast cancer via Hedgehog/Gli1 signaling pathway. J Ethnopharmacol, 2022. 283 : p. 114689. Zhang, W., W. Liu, and X. Hu, Robinin inhibits pancreatic cancer cell proliferation, EMT and inflammation via regulating TLR2-PI3k-AKT signaling pathway. Cancer Cell Int, 2023. 23 (1): p. 328. Gupta, N., et al., Notch3 induces epithelial-mesenchymal transition and attenuates carboplatin-induced apoptosis in ovarian cancer cells. Gynecol Oncol, 2013. 130 (1): p. 200-6. Arrindell, J. and B. Desnues, Vimentin: from a cytoskeletal protein to a critical modulator of immune response and a target for infection. Front Immunol, 2023. 14 : p. 1224352. Ivaska, J., et al., Novel functions of vimentin in cell adhesion, migration, and signaling. Exp Cell Res, 2007. 313 (10): p. 2050-62. Li, C. and Y.Q. Ma, Prognostic significance of sex determining region Y-box 2, E-cadherin, and vimentin in esophageal squamous cell carcinoma. World J Clin Cases, 2022. 10 (27): p. 9657-9669. Quan, J., A.M. Bode, and X. Luo, ACSL family: The regulatory mechanisms and therapeutic implications in cancer. Eur J Pharmacol, 2021. 909 : p. 174397. Magtanong, L., et al., Exogenous Monounsaturated Fatty Acids Promote a Ferroptosis-Resistant Cell State. Cell Chem Biol, 2019. 26 (3): p. 420-432.e9. Saliakoura, M., et al., The ACSL3-LPIAT1 signaling drives prostaglandin synthesis in non-small cell lung cancer. Oncogene, 2020. 39 (14): p. 2948-2960. Yin, L., et al., HOOK1 Inhibits the Progression of Renal Cell Carcinoma via TGF-β and TNFSF13B/VEGF-A Axis. Adv Sci (Weinh), 2023. 10 (17): p. e2206955. Chen, K., et al., MAP4K4 promotes ovarian cancer metastasis through diminishing ADAM10-dependent N-cadherin cleavage. Oncogene, 2023. 42 (18): p. 1438-1452. Delpire, E., The mammalian family of sterile 20p-like protein kinases. Pflugers Arch, 2009. 458 (5): p. 953-67. Sun, R., et al., Annexin10 promotes extrahepatic cholangiocarcinoma metastasis by facilitating EMT via PLA2G4A/PGE2/STAT3 pathway. EBioMedicine, 2019. 47 : p. 142-155. Kaufmann, R., et al., Thrombin-mediated hepatocellular carcinoma cell migration: cooperative action via proteinase-activated receptors 1 and 4. J Cell Physiol, 2007. 211 (3): p. 699-707. Mou, Y., et al., Ferroptosis, a new form of cell death: opportunities and challenges in cancer. J Hematol Oncol, 2019. 12 (1): p. 34. Gabbs, M., et al., Advances in Our Understanding of Oxylipins Derived from Dietary PUFAs. Adv Nutr, 2015. 6 (5): p. 513-40. Yang, X., et al., miR-18a promotes glioblastoma development by down-regulating ALOXE3-mediated ferroptotic and anti-migration activities. Oncogenesis, 2021. 10 (2): p. 15. Endo, M., The Roles of ANGPTL Families in Cancer Progression. J uoeh, 2019. 41 (3): p. 317-325. Yang, W.H., et al., A TAZ-ANGPTL4-NOX2 Axis Regulates Ferroptotic Cell Death and Chemoresistance in Epithelial Ovarian Cancer. Mol Cancer Res, 2020. 18 (1): p. 79-90. Zhuang, W., et al., An immunogenomic signature for molecular classification in hepatocellular carcinoma. Mol Ther Nucleic Acids, 2021. 25 : p. 105-115. Yunna, C., et al., Macrophage M1/M2 polarization. Eur J Pharmacol, 2020. 877 : p. 173090. Wang, Y., et al., FOXO1 promotes tumor progression by increased M2 macrophage infiltration in esophageal squamous cell carcinoma. Theranostics, 2020. 10 (25): p. 11535-11548. Chen, L., et al., Aberrant epithelial cell interaction promotes esophageal squamous-cell carcinoma development and progression. Signal Transduct Target Ther, 2023. 8 (1): p. 453. Song, J., et al., A Novel Ferroptosis-Related Biomarker Signature to Predict Overall Survival of Esophageal Squamous Cell Carcinoma. Front Mol Biosci, 2021. 8 : p. 675193. Additional Declarations No competing interests reported. Supplementary Files supplementraytable1.docx Supplementary material Supplementary Table 1. Chi-square test of the immune infiltration abundance in different risk groups in GEO cohort Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6456491","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":452056179,"identity":"1efce2cb-8302-42b8-989d-be64e82c9fd1","order_by":0,"name":"Zhidong Wang","email":"","orcid":"","institution":"The Third Affiliated Hospital of Soochow University","correspondingAuthor":false,"prefix":"","firstName":"Zhidong","middleName":"","lastName":"Wang","suffix":""},{"id":452056180,"identity":"85487eec-588e-4afa-a635-dc3fb1ba0670","order_by":1,"name":"Cheng Gong","email":"","orcid":"","institution":"The Third Affiliated Hospital of Soochow University","correspondingAuthor":false,"prefix":"","firstName":"Cheng","middleName":"","lastName":"Gong","suffix":""},{"id":452056181,"identity":"b5d54125-f4d3-412c-b542-3f7e06e2e266","order_by":2,"name":"Ce Chao","email":"","orcid":"","institution":"The Third Affiliated Hospital of Soochow University","correspondingAuthor":false,"prefix":"","firstName":"Ce","middleName":"","lastName":"Chao","suffix":""},{"id":452056182,"identity":"9f9e891e-e8ef-45ef-a090-c826196c4401","order_by":3,"name":"Youpu Zhang","email":"","orcid":"","institution":"The Third Affiliated Hospital of Soochow University","correspondingAuthor":false,"prefix":"","firstName":"Youpu","middleName":"","lastName":"Zhang","suffix":""},{"id":452056183,"identity":"9713c2d8-5c13-408d-8b77-22109daa2659","order_by":4,"name":"Yiongxiang Qian","email":"","orcid":"","institution":"The Third Affiliated Hospital of Soochow University","correspondingAuthor":false,"prefix":"","firstName":"Yiongxiang","middleName":"","lastName":"Qian","suffix":""},{"id":452056184,"identity":"05d30d2d-4b91-4ff5-b72d-a80c97035c68","order_by":5,"name":"Min Wang","email":"","orcid":"","institution":"The Third Affiliated Hospital of Soochow University","correspondingAuthor":false,"prefix":"","firstName":"Min","middleName":"","lastName":"Wang","suffix":""},{"id":452056185,"identity":"209b1ed7-5874-4786-b1e8-ded7effb8057","order_by":6,"name":"Bin Wang","email":"","orcid":"","institution":"The Third Affiliated Hospital of Soochow University","correspondingAuthor":false,"prefix":"","firstName":"Bin","middleName":"","lastName":"Wang","suffix":""},{"id":452056186,"identity":"16d64535-7a68-4793-9929-f1b97ccb23eb","order_by":7,"name":"Yang Liu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyElEQVRIiWNgGAWjYFCCBIMDFTYSDAzsjY0PPhCt5UwaUAvP4WbDGcRqYTiTBqQl0tukOYjRwM+evPHAgQQLOfOZDxukGRjs5HQbCGiR7HlWANQiYSxzO7HBuIAh2djsAAEtBjdyDA5//CGROEM6sSF5BsOBxG2EtNgDtYBsqZ8hebDhMA8xWgwkIFoSJCQYG5uJ0iJxBuIXwxk8ic2MMwyI8At/e/LmDwcS6uQl2I8///Ghwk6OoBZ0d5KmfBSMglEwCkYBDgAAt2NIRJLKsnoAAAAASUVORK5CYII=","orcid":"","institution":"The Third Affiliated Hospital of Soochow University","correspondingAuthor":true,"prefix":"","firstName":"Yang","middleName":"","lastName":"Liu","suffix":""}],"badges":[],"createdAt":"2025-04-15 15:53:23","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6456491/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6456491/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":82355510,"identity":"62755836-b742-4dcc-a7ab-f1b7f66f89ff","added_by":"auto","created_at":"2025-05-09 11:14:48","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":419327,"visible":true,"origin":"","legend":"\u003cp\u003eStudy flow chart.\u003c/p\u003e","description":"","filename":"figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-6456491/v1/c30347abf11a7fcbdec1e3d9.png"},{"id":82355511,"identity":"9ed6f251-db20-4eb0-8ea8-5091d5dbed13","added_by":"auto","created_at":"2025-05-09 11:14:48","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":11464035,"visible":true,"origin":"","legend":"\u003cp\u003e(A) A volcanic plot of DEGs. (B) A Venn diagram showing that 714 differentially expressed FRGs and ERGs were identified in the GEO cohort. (C-D) LASSO regression was performed to generate the crucial genes and the correlation coefficient of genes. (E) A chord diagram displaying the interconnectivity among FRGs and ERGs. The thickness and color of the ribbons between FRGs and ERGs correlate to the correlation of genes expression. (F, G) Distribution and survival status of patents based on FEIS in the GEO cohort. (H) The Kaplan–Meier curves of high-FEIS group and low-FEIS group in the GEO cohort. (I) PCA plots of the GEO cohort based on FEIS. (J) ROC curves demonstrated the predictive efficiency of the FEIS in the GEO cohort. (K, L) Distribution and survival status of patents based on FEIS in the TCGA cohort. (M) The Kaplan–Meier curves of high-FEIS group and low-FEIS group in the TCGA cohort. (N) PCA plots of the TCGA cohort based on FEIS. (O) ROC curves demonstrated the predictive efficiency of the FEIS in the TCGA cohort. *p\u0026lt;0.05; **p\u0026lt;0.01; ***p\u0026lt;0.001; ****p\u0026lt;0.0001; ns, not significant.\u003c/p\u003e","description":"","filename":"figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-6456491/v1/633ecc5559939ca8d41c9a3c.png"},{"id":82357657,"identity":"cca7c753-6dd3-4385-8bc2-7dccf8f87867","added_by":"auto","created_at":"2025-05-09 11:22:48","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":5877363,"visible":true,"origin":"","legend":"\u003cp\u003eSelection of prognostic related signatures. (A) A heatmap showed the relationship between FEIS and TNM stage in the GEO cohort. (B-E) Univariate and multivariate Cox regression analyses in the GEO and TCGA cohorts, and the results showed that FEIS was an independent prognostic factor in tow cohorts respectively. *p\u0026lt;0.05; **p\u0026lt;0.01; ***p\u0026lt;0.001; ****p\u0026lt;0.0001; ns, not significant.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-6456491/v1/a72e3700282e5fe356c9ad50.png"},{"id":82355519,"identity":"138e02ee-5785-434a-844c-d2ef021762a2","added_by":"auto","created_at":"2025-05-09 11:14:48","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":15176883,"visible":true,"origin":"","legend":"\u003cp\u003eExpression of SOFs and EMT markers in TCGA and GEO cohorts, and enrichment analysis in the high-FEIS group. (A, C) Box plots showing the expression profiles of SOFs in the GEO and TCGA cohorts. (B, D) Box plots showing the expression profiles of EMT markers in the GEO and TCGA cohorts. (E) Pathways associated with EMT enriched in the high-FEIS group in the GEO cohort. (F) Pathways associated with EMT enriched in the high-FEIS group in the TCGA cohort. *p\u0026lt;0.05; **p\u0026lt;0.01; ***p\u0026lt;0.001; ****p\u0026lt;0.0001; ns, not significant.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-6456491/v1/a1e49fe7f6de5d95c2786139.png"},{"id":82359717,"identity":"c73521a1-8f21-443f-b48d-523bfb83db65","added_by":"auto","created_at":"2025-05-09 11:30:49","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":24898243,"visible":true,"origin":"","legend":"\u003cp\u003eimmunoinfiltration analysis. (A) A heatmap of ssGSEA in the GEO cohort showing the high immune infiltration cluster (cluster1) and the low immune infiltration cluster(cluster2). (B) Comparison between the fractions of immune cells in the high-FEIS and low-FEIS groups of the GEO cohort via the CIBERSORT method. (C-G) Differential expression of immune checkpoints between high- and low-FEIS group in GEO cohort. *p\u0026lt;0.05; **p\u0026lt;0.01; ***p\u0026lt;0.001; ****p\u0026lt;0.0001; ns, not significant.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-6456491/v1/110ed23b084d94046f1b4f02.png"},{"id":82355515,"identity":"b19a115c-b9e4-464f-85dd-7c073dce56be","added_by":"auto","created_at":"2025-05-09 11:14:48","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":5901538,"visible":true,"origin":"","legend":"\u003cp\u003ePrognostic nomogram establishment and validation. Prognostic nomogram establishment and validation. (A) Nomogram based on FEIS and TNM stage. (B-D) ROC curves for the nomogram, FEIS, and TNM stage in the GEO cohort at 1-, 3-, and 5-year. (E, F) ROC curves for the nomogram, FEIS, and TNM stage in the TCGA cohort at 1- and 2- year. (G-I) Calibration curves of 1-, 3- and 5-year overall survival for patients in GEO cohort. (J, K) Calibration curves of 1-, 2-year overall survival for patients in TCGA cohort. (L-N) Decision curve analysis of the nomogram, FEIS, and TNM stage for survival prediction of patient in the GEO cohort at 1-, 3-, and 5-year. (O, P) Decision curve analysis of the nomogram, FEIS, and TNM stage for survival prediction of patient in the TCGA cohort at 1- and 2-year. *p\u0026lt;0.05; **p\u0026lt;0.01; ***p\u0026lt;0.001; ****p\u0026lt;0.0001; ns, not significant.\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-6456491/v1/5a38453a596404eeb4f002c7.png"},{"id":84300570,"identity":"7b8b1788-cb0d-45e9-a995-a09ee30b0fb1","added_by":"auto","created_at":"2025-06-10 10:32:47","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":42607322,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6456491/v1/5117a879-0bf0-43c7-b22c-5644699a83f9.pdf"},{"id":82355505,"identity":"7c5adb51-65f2-4661-8804-1168c232e2ff","added_by":"auto","created_at":"2025-05-09 11:14:48","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":15985,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary material\u003c/p\u003e\n\u003cp\u003eSupplementary Table 1. Chi-square test of the immune infiltration abundance in different risk groups in GEO cohort\u003c/p\u003e","description":"","filename":"supplementraytable1.docx","url":"https://assets-eu.researchsquare.com/files/rs-6456491/v1/9170e8830a80fce6de70eaa3.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"A dual EMT-ferroptosis gene signature predicts survival and immune infiltration in esophageal squamous cell carcinoma","fulltext":[{"header":"Introduction","content":"\u003cp\u003eEsophageal cancer (EC) is the seventh most common cancer and the sixth leading cause of cancer-related deaths worldwide[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Esophageal squamous cell carcinoma (ESCC) is a prominent subtype of EC, and its adverse prognosis stems from its subtle initial symptoms, propensity for metastasis, resistance to radiotherapy, and tendency for tumor recurrence [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. In recent years, there has been considerable progress in advancing multidisciplinary and surgical therapies for ESCC. However, despite these advancements, the median survival rate remains low for ESCC patients [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Hence, it is essential to search for new biomarkers to promote personalized treatment and improve survival rate and quality of life of ESCC patients.\u003c/p\u003e \u003cp\u003eFerroptosis is a specific cell death program associated with the accumulation of iron ions and uncontrolled oxidative reactions within a cell [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. There has been an exponential growth in research on ferroptosis, with numerous studies highlighting its remarkable role in cancer advancement and treatment outcomes [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Despite extensive efforts to develop novel therapeutic strategies, ESCC treatment presents several complex challenges, including increased chemotherapy drug resistance and reduced sensitivity of ESCC cells to radiotherapy. A study showed that STC2, as a promising therapeutic target, addressed ESCC radioresistance by inactivating PRMT5 to prevent DNA damage repair and inhibition of ferroptosis [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Gong et al. found that targeting metabolic checkpoints could be used as an effective strategy for ESCC therapy by suppressing tumor ferroptosis [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Additionally, another study showed that treatment with 5-aminolevulinic acid suppressed the GPX4 and HMOX1 overexpression, thereby facilitating ferroptosis in ESCC cells [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Hence, focusing on ferroptosis genes and their related pathways could advance research on suppressing tumor growth. Understanding ferroptosis status in various ESCC patients is therefore crucial to improve the effectiveness of personalized precision treatments.\u003c/p\u003e \u003cp\u003eEpithelial-mesenchymal transition (EMT) is a pivotal cellular program essential for embryogenesis, wound healing, and malignancy progression [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. In neoplasms, EMT endows cancer cells with high tumor-initiating and metastatic capabilities and also influences the growth of cancer stem cells, tumor immune evasion, and resistance to drugs [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. A previous study showed that EMT and cancer cell stemness were associated with chemotherapy resistance of EC, and addressing both these properties could serve as an improved approach to combat drug resistance in EC [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Research on EMT-like modifications in tumor cells indicates enhanced aggressiveness, with a strong association between EMT transcriptome characteristics and unfavorable prognoses across various cancer patient cohorts [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Considering these characteristics, the complex biological mechanism of EMT is thought to be closely associated with carcinogenesis, and directing interventions toward EMT pathways has emerged as a promising approach for ESCC therapy.\u003c/p\u003e \u003cp\u003eThe potential association between ferroptosis and EMT, both of which are critical biological processes in cancer cell, is attracting increasing attention. According to previous research, the occurrence of EMT in cancer cells increases their susceptibility to ferroptosis. Bach1, a crucial regulator of glioma invasion, orchestrates various ECM-related functions, and the overexpression of Bach1 reduces ferroptosis in glioma cells [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. In melanoma cells, TGF-β1 triggers EMT and simultaneously elevates lipid peroxidation levels, thereby promoting the cellular process of ferroptosis[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. Another noteworthy finding is that ferroptosis may increase EMT inhibition. Guan et al. reported that DpdtbA exerted its inhibitory effect on EMT by activating the p53 and PHD2/HIF-1α pathways [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. However, Yao et al. revealed that the inclination toward ferroptosis positively influenced EMT occurrence in the epithelial cells of lung adenocarcinoma [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIncreasing evidence indicates that the subtle interaction between ferroptosis and EMT plays a substantial role in the biological processes of tumors. Hence, cancer treatment efforts should be directed toward targeting ferroptosis and EMT. In the present research, we are committed to developing a prognostic model that comprehensively considers the impact of ferroptosis and EMT on the prognosis of ESCC.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData collection and acquisition of gene expression data\u003c/h2\u003e \u003cp\u003eThe dataset GSE53625, derived from the GEO database, served as the training set for the analysis. The dataset was accessed via the \u0026ldquo;GEOquery\u0026rdquo; package in R, and the \u0026ldquo;limma\u0026rdquo; package was employed to process the mRNA expression matrix. This dataset comprised 179 ESCC carcinoma specimens and 179 adjacent healthy tissue specimens. The TCGA database was screened to obtain complete clinical information of the patients as the validation set. After excluding samples with missing data, 94 samples were eligible for the analysis, which included 91 ESCC specimens and 3 adjacent normal tissue specimens. Thus, the gene expression data were obtained from a total of 270 tumor samples and 182 normal tissue samples, accompanied with clinicopathological information (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The gene expression data were subjected to normalization with log2(TPM\u0026thinsp;+\u0026thinsp;1). We obtained 259 ferroptosis-related genes (FRGs) from the FerrDb database and 1011 protein-coding EMT-related genes (ERGs) from the dbEMT 2.0 database.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBaseline characteristics of ESCC patients in GEO and TCGA cohorts\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGEO\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eTCGA\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;179)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;91)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e59\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e32\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e146\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e77\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTNM Stage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI-II\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e87\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e61\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIII-IV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTobacco use\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e114\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAlcohol use\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e106\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e73\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTumor loation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLower\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMiddle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e97\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUpper\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTumor grade\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePoorly\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModerately\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWell\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eIdentification of critical genes\u003c/h3\u003e\n\u003cp\u003eSignificantly different genes (SDGs) were screened out by comparing tumor samples with normal samples. The cutoff values were set as false discovery rate (FDR)\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and |log2FC| \u0026gt; 0.2. Then FRGs and ERGs were then intersected with SDGs, and the overlapping genes were shown with a Venn diagram. Subsequently, univariate Cox regression analysis was conducted to discover potential biomarkers. LASSO regression analysis was employed to simplify the model and ruduce overfitting risk. The \u0026ldquo;glmnet\u0026rdquo; package was used to determine the key genes and to generate the corresponding coefficients for these genes. Finally, the ferroptosis and EMT integrated score (FEIS) was determined with the following equation: FEIS =\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\sum\\:_{i=1}^{n}{coef}_{i}*{geneExp}_{i}\\)\u003c/span\u003e\u003c/span\u003e, where the \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{coef}_{i}\\)\u003c/span\u003e\u003c/span\u003e represents the coefficient and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{geneExp}_{i}\\)\u003c/span\u003e\u003c/span\u003e denotes the expression value of each key gene. Based on the median FEIS value, all patients were categorized into high-FEIS group or low-FEIS group. Relationship between key FRGs and ERGs calculated using Pearson correlation coefficients based on gene expression values, and the chord diagram provides an intuitive visualization.\u003c/p\u003e\n\u003ch3\u003eEnrichment analysis\u003c/h3\u003e\n\u003cp\u003eGene Set Enrichment Analysis (GSEA) was employed to compare the enrichment of gene sets between the different risk groups. We utilized GSEA software (version 4.3.2) to conduct enrichment analysis, and selected \u0026ldquo;c2.cp.kegg.v7.5.1.symbols.gmt'\u0026rdquo; as the gene sets database. The number of permutations was set to 1,000, with the phenotype labels defined as \u0026ldquo;low FEIS versus high FEIS,\u0026rdquo; and the enrichment statistic set to \u0026ldquo;weighted.\u0026rdquo; To avoid dataset specificity, the same approach using the TCGA dataset. Ensuring the significance and reliability of the results, we applied the following criteria for pathway enrichment: nominal p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and FDR\u0026thinsp;\u0026lt;\u0026thinsp;0.25.\u003c/p\u003e\n\u003ch3\u003eImmunoinfiltration analysis\u003c/h3\u003e\n\u003cp\u003eThe single-sample Gene Set Enrichment Analysis (ssGSEA) method, implemented using the R package \u0026ldquo;GSVA\u0026rdquo;, was applied to calculate and analyze the enrichment scores of immune-related gene sets in each sample. We performed cluster analysis, dividing the samples into high immune infiltration and low immune infiltration groups based on the clustering results. Additionally, a validated expression matrix of 22 immune cell subtypes (LM22) was utilized to analyze the proportions of immune cells in tumor tissues.\u003c/p\u003e\n\u003ch3\u003eConstruction and validation of the nomogram\u003c/h3\u003e\n\u003cp\u003eWe constructed a nomogram to forecast the clinical outcomes for ESCC patients in the TCGA and GEO databases. TNM stage and FEIS were included as the prognostic factors. Receiver operating characteristic (ROC) curves were generated to validate the constructed nomogram.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eStatistical analysis was performed using GraphPad (version 9.5.1) and R software (version 4.3.0). Categorical variables were compared using the chi-square test. The levels of gene expression, immune cell infiltration, and immune pathway activation between the different subgroups were analyzed using Wilcoxon test. Visualization was performed with \u0026ldquo;rms,\u0026rdquo; \u0026ldquo;ggbiplot,\u0026rdquo; \u0026ldquo;pheatmap,\u0026rdquo; \u0026ldquo;ggplot2,\u0026rdquo; \u0026ldquo;survival,\u0026rdquo; and \u0026ldquo;ggpubr\u0026rdquo; packages. p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant.\u003c/p\u003e \u003c/div\u003e"},{"header":"Result","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eBiomarker identification and survival analysis\u003c/h2\u003e \u003cp\u003eAs shown in the flowchart (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), this study included 1208 FRGs and ERGs. Variations in gene expression were exhibited by a total of 714 genes between adjacent normal tissues and malignant tissues (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, B). LASSO regression analysis indicated nine genes associated with ferroptosis and EMT (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC, D). These included 6 ERGs (\u003cem\u003eVIM\u003c/em\u003e, \u003cem\u003eHOOK1\u003c/em\u003e, \u003cem\u003eMAP4K4\u003c/em\u003e, \u003cem\u003ePLA2G4A\u003c/em\u003e, \u003cem\u003eF2RL2\u003c/em\u003e, and \u003cem\u003eLASP1\u003c/em\u003e) and 3 FRGs (\u003cem\u003eACSL3\u003c/em\u003e, \u003cem\u003eALOXE3\u003c/em\u003e, and \u003cem\u003eANGPTL7\u003c/em\u003e). The chord diagram visually represented FRGs in the green module and ERGs in the pink module. The thickness and color of the ribbons indicated gene expression correlation between these categories, suggesting their interaction (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE). Patients were assigned to the high- and low-FEIS groups based on the midpoint FEIS value (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF), with those in the high-FEIS group showing poorer prognosis and shorter survival duration compared to the low-FEIS group (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG, H). According to principal component analysis (PCA), the patient composition differed significantly between the two groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eI). The areas under the ROC curve (AUC) values at 1-, 3-, and 5- year survival period was 0.737, 0.756, and 0.771, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eJ).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eModel validation in the TCGA cohort\u003c/h2\u003e \u003cp\u003eThe TCGA database cohort was used as a validation set. In line with findings from the GEO cohort, the high-FEIS group showed adverse prognosis and increased number of early deaths (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eK-M). The results of PCA showed clear separation of the patients into two distinct subgroups (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eN). The AUC values at 1- and 2-year survival period were 0.638 and 0.764, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eO).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eSubgroup analysis for the prognostic signature\u003c/h2\u003e \u003cp\u003eAs shown in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, \u003cem\u003eHOOK1\u003c/em\u003e, \u003cem\u003ePLA2G4A\u003c/em\u003e, and \u003cem\u003eALOXE3\u003c/em\u003e exhibited protective effects on ESCC patients (hazard ratio [HR]\u0026thinsp;\u0026lt;\u0026thinsp;1, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), while \u003cem\u003eACSL3\u003c/em\u003e, \u003cem\u003eVIM\u003c/em\u003e, \u003cem\u003eF2RL2\u003c/em\u003e, \u003cem\u003eANGPTL7\u003c/em\u003e, \u003cem\u003eMAP4K4\u003c/em\u003e, and \u003cem\u003eLASP1\u003c/em\u003e were identified as the risk factors (HR\u0026thinsp;\u0026gt;\u0026thinsp;1, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01). The gene expression levels were shown in the heatmap for the subgroups within the GEO cohort (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). The baseline clinical data between the high- and low-FEIS groups were analyzed, revealing that the high-FEIS group had a significantly greater proportion of patients with stage III-IV cancer compared to the low-FEIS group (p\u0026thinsp;\u0026lt;\u0026thinsp;0.005) (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Next, the FEIS and TNM stage were included in the univariate Cox analysis to assess their effects on survival time. According to the forest plot, both the FEIS and TNM stage were influential indicators for predicting patient outcomes (GEO cohort: FEIS, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; TNM stage, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01; TCGA cohort: FEIS, p\u0026thinsp;=\u0026thinsp;0.003; TNM stage, p\u0026thinsp;=\u0026thinsp;0.013; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB, D). In the multivariate Cox regression analysis, the results indicated that the FEIS was an independent prognostic factor for ESCC patients in both GEO (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, HR\u0026thinsp;=\u0026thinsp;2.775; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC) and TCGA cohorts (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, HR\u0026thinsp;=\u0026thinsp;2.631; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe 9 genes in the LASSO model\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGene\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHR 95L\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eHR 95H\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHOOK1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.761353\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.636666635\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.910458363\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.002808\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePLA2G4A\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.802401\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.696918162\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.923849755\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.002203\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eALOXE3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.820369\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.716824994\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.938869542\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.004024\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eF2RL2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.273074\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.093504583\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.482131596\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.001857\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eANGPTL7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.331697\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.103461726\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.607139998\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.002823\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMAP4K4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.48965\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.105592942\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.007119505\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.008797\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eACSL3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.642826\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.200264165\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.248570269\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.001936\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVIM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.756214\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.22448229\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.518849046\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.002209\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLASP1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.974352\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.195652903\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.260196908\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.007854\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eClinicopathological features of ESCC patients in low- and high-FEIS group\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eGEO\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e \u003cp\u003eTCGA\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eCharacteristics\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLow-FEIS\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHigh-FEIS\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eP value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eLow-FEIS\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eHigh-FEIS\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eP value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e(N\u0026thinsp;=\u0026thinsp;90)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(N\u0026thinsp;=\u0026thinsp;89)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003e(N\u0026thinsp;=\u0026thinsp;46)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003e(N\u0026thinsp;=\u0026thinsp;45)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAge\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.125\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026le;\u0026thinsp;60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e50 (55.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e49 (55.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e26 (56.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e33 (73.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e40 (44.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e40 (44.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e20 (43.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e12 (26.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eGender\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.773\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e75 (83.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e71 (79.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e38 (82.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e39 (86.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e15 (16.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e18 (20.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e8 (17.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e6 (13.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eTNM Stage\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.027\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI-II\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e52 (57.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e35 (39.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e36 (78.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e25 (55.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIII-IV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e38 (42.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e54 (60.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e10 (21.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e20 (44.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eTobacco use\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e28 (31.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e37 (41.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e62 (68.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e52 (58.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAlcohol use\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e33 (36.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e40 (44.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e57 (63.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e49 (55.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003etumor location\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLower\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e33 (36.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e29 (32.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMiddle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e46 (51.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e51 (57.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUpper\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e11 (12.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e9 (10.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003etumor grade\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModerately\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e55 (61.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e43 (48.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePoorly\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e19 (21.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e30 (33.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWell\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e16 (17.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e16 (18.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eAnalysis of ferroptosis and EMT status\u003c/h2\u003e \u003cp\u003e \u003cem\u003eAR, PDK4, MEF2C, ACSL3, USP11, NCOA3, PTPN18, TMSB4X, SLC40A1\u003c/em\u003e, and \u003cem\u003eSIRT1\u003c/em\u003e are known suppressors of ferroptosis (SOFs)[\u003cspan additionalcitationids=\"CR21 CR22 CR23 CR24 CR25 CR26 CR27 CR28\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. We compared the expression patterns of these SOFs between the two subgroups to evaluate ferroptosis status. In the GEO cohort, the expression levels of the \u003cem\u003eAR\u003c/em\u003e, \u003cem\u003ePDK4\u003c/em\u003e, \u003cem\u003eMEF2C\u003c/em\u003e, \u003cem\u003eACSL3\u003c/em\u003e, \u003cem\u003eUSP11\u003c/em\u003e, \u003cem\u003eNCOA3\u003c/em\u003e, \u003cem\u003ePTPN18\u003c/em\u003e, \u003cem\u003eTMSB4X\u003c/em\u003e, \u003cem\u003eSLC40A1\u003c/em\u003e, and \u003cem\u003eSIRT1\u003c/em\u003e genes were significantly increased in the high-FEIS group (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). Similarly, in the TCGA cohort, the high-FEIS group showed a significant increase in the expression levels of the \u003cem\u003eMEF2C\u003c/em\u003e, \u003cem\u003eUSP11\u003c/em\u003e, \u003cem\u003ePTPN18\u003c/em\u003e, \u003cem\u003eTMSB4X\u003c/em\u003e, \u003cem\u003eSLC40A1\u003c/em\u003e, and \u003cem\u003eSIRT1\u003c/em\u003e genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). These findings indicated a potentially suppressive ferroptosis status in the high-FEIS group.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eSubsequently, we further analyzed the expression levels of EMT markers within the subgroups. In the GEO cohort, the expression levels of the \u003cem\u003eZEB1\u003c/em\u003e, \u003cem\u003eTWIST1\u003c/em\u003e, \u003cem\u003eVIM\u003c/em\u003e, \u003cem\u003eFN1\u003c/em\u003e, \u003cem\u003eZEB2\u003c/em\u003e, \u003cem\u003eFOXC2\u003c/em\u003e, \u003cem\u003ePTX3\u003c/em\u003e, \u003cem\u003eWT1\u003c/em\u003e, \u003cem\u003eSEMA3E\u003c/em\u003e, \u003cem\u003eNKX3-2\u003c/em\u003e, \u003cem\u003eCYP1B1\u003c/em\u003e, \u003cem\u003eNKX6-1\u003c/em\u003e, \u003cem\u003eSCUBE3\u003c/em\u003e, \u003cem\u003eAGTR1\u003c/em\u003e, and \u003cem\u003eBVES\u003c/em\u003e genes were significantly upregulated in the high-FEIS group (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). In the TCGA cohort, the high-FEIS group showed a notable increase in the gene expression levels of the relevant EMT markers, except for \u003cem\u003eSCUBE3\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD). These results suggest that the high-FEIS group patients had more active expression of EMT-related genes, thus making them more prone to adverse prognostic events such as tumor metastasis. The GSEA analysis results further confirmed that biological pathways related to EMT were abundant in the high-FEIS group in the GEO and TCGA cohort. These pathways included the overexpression of chondroitin sulfate and actin cytoskeleton, and the enrichment of focal junction and EMC (Extracellular Matrix) receptor interaction pathways in the GEO cohort (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE). Similarly, the pathways related to focal junction and EMC receptor interaction were enriched in the high FEIS group in the TCGA cohort (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF). Additionally, the Hedgehog, Notch3 and PI3K/AKT signaling pathways, which could promote the occurrence of EMT, were upregulated in the high-FEIS group in the TCGA cohort[\u003cspan additionalcitationids=\"CR31\" citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e] (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF).\u003c/p\u003e \u003cp\u003e \u003cb\u003eTumor immune microenvironment in the subgroups.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eAccording to the results of ssGSEA, cluster analysis was performed to divide the samples into high immune infiltration and low immune infiltration clusters (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). A chi-square test was conducted to determine whether the high-FEIS group showed elevated or reduced levels of immune cell infiltration. The results indicated a significant prevalence of high immune infiltration levels in the high-FEIS group (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05; Supplementary Table\u0026nbsp;1). According to the CIBERSORT analysis, CD8 T cells, activated CD4 memory T cells, M1 macrophages, and M2 macrophages showed relatively higher abundance in the high-FEIS group. The infiltration levels of monocytes, activated dendritic cells, and activated mast cells were higher in the low-FEIS group (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). Further comparison of the expression levels of immune checkpoint molecules between the subgroups revealed a notable increase in the expression levels of the \u003cem\u003eBTLA\u003c/em\u003e, \u003cem\u003eCTLA4\u003c/em\u003e, \u003cem\u003eICOSLG\u003c/em\u003e, \u003cem\u003ePDCD1LG2\u003c/em\u003e, and \u003cem\u003eTNFRSF9\u003c/em\u003e genes in the high-FEIS group (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC-G).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eEstablishment and validation of the prognostic nomogram\u003c/h2\u003e \u003cp\u003eWe selected the FEIS and TNM stage as prognostic factors to construct a nomogram and evaluated its precision and clinical effectiveness in the GEO and TCGA datasets (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). Based on the time-dependent ROC curves, the AUC values of the diagnostic model were 0.727 (1-year), 0.779 (3-year), and 0.764 (5-year) in the GEO cohort (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB-D). In the TCGA cohort, because few patients survived beyond 3 years, the AUC values were 0.696 and 0.782 at 1 and 2 years of survival (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE, F). These values indicated that the developed nomogram had reasonable sensitivity and specificity in the GEO and TCGA cohort. The calibration curve showed a moderate consistency between the predicted values of this diagnostic model and the actual values in the GEO (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eG-I) and TCGA cohorts (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eJ, K). DCA (decision curve analysis) indicated that the developed nomogram had dequate clinical utility in both GEO cohort (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eL-N) and TCGA cohort (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eO, P).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis research comprehensively analyzed how ferroptosis and EMT impact the prognosis for ESCC patients and constructed a predictive nomogram model that incorporated biomarkers related to both these processes.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOur proposed signature comprised 6 ERGs (\u003cem\u003eVIM\u003c/em\u003e, \u003cem\u003eHOOK1\u003c/em\u003e, \u003cem\u003eMAP4K4\u003c/em\u003e, \u003cem\u003ePLA2G4A\u003c/em\u003e, \u003cem\u003eF2RL2\u003c/em\u003e, and \u003cem\u003eLASP1\u003c/em\u003e) and 3 FRGs (\u003cem\u003eACSL3\u003c/em\u003e, \u003cem\u003eALOXE3\u003c/em\u003e, and \u003cem\u003eANGPTL7\u003c/em\u003e). These genes are highly associated with the onset and development of cancers. \u003cem\u003eVIM\u003c/em\u003e is crucial in various biological processes and diseases. This gene encodes the vimentin protein that regulates adhesion among cells and participates in migration and invasion processes[33, 34]. Chao et al. demonstrated that the sex-determining region Y-Box 2 regulated the expression of VIM and the activity of signaling pathways associated with EMT, thereby influencing the phenotypic transition and functional characteristics of cells, which subsequently affects the invasion and metastatic capability of ESCC tumor cells [35]. ACSL3 participates in regulating lipid metabolism and cellular energy balance[36]. Magtanong et al. found that ACSL3 could activate the exogenous monounsaturated fatty acids to reduce the sensitivity pertaining to lethal oxidation in plasma membrane lipids, thereby enhancing cellular resistance to ferroptosis [37]. \u003cem\u003eACSL3\u003c/em\u003e expression is upregulated across various cancer types, and it predominantly facilitates cancer development through diverse mechanisms. Another study indicated that ACSL3 enhanced the growth and metastatic potential of cancer cells [38].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn the present study, we observed that Hook1 was a protective factor for EC patients; this finding was consistent with a previous report that HOOK1 could suppress renal cell carcinoma progression through the TGF-\u0026beta; and TNFSF13B/VEGF-A pathway [39]. \u003cem\u003eMAP4K4\u003c/em\u003e encodes a serine/threonine kinase that plays a crucial role in cellular signal transduction pathways and promotion of ovarian cancer metastasis [40, 41].\u003cem\u003e\u0026nbsp;PLA2G4A\u003c/em\u003e is a calcium-dependent phospholipase A2 family member involved in cell adhesion, migration, regulation of cell cycle progression, and various physiological processes related to tumor initiation and development. A previous study noted that annexin A10 facilitates metastasis of extrahepatic cholangiocarcinoma by promoting EMT through the PLA2G4A/PGE2/STAT3 pathway [42]. The F2RL2 protein, similar to F2RL3, plays several important roles in cell and tumor biology. A previous study showed that F2RL3 enhanced the migration and invasion abilities of tumor cells by activating different signaling pathways, facilitating their passage through the cell matrix and spread to other tissues and organs, thereby promoting tumor metastasis [43].\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eThe gene\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cem\u003eALOXE3\u003c/em\u003e encodes the lipoxygenase protein. Lipoxins are the products of the lipoxygenase catalytic reaction, which play critical roles in cellular signaling, inflammation, immune responses, and other physiological processes. An excessive amount of lipoxins may promote the occurrence of ferroptosis [44, 45].\u0026nbsp;According to a previous study,\u0026nbsp;the absence of ALOXE3 leads to the onset of malignant glioblastoma, while the inhibition of \u003cem\u003eALOXE3\u003c/em\u003e expression confers resistance to p53-mediated ferroptosis in glioblastoma cells [46]. Angiopoietin-like proteins (ANGPTLs), with a function in chronic inflammation, are one of the key factors in carcinogenesis, and the proinflammatory factor ANGPTL7 regulates cancer progression [47]. Although the role of ANGPTL7 in ferroptosis is unclear, a previous study reported that the homologous protein ANGPTL3 could regulate the activity of ferroptosis mechanisms and the response of tumor cells to chemotherapeutic drugs in epithelial ovarian cancer through the TAZ-ANGPTL4-NOX2 axis\u0026nbsp;[48].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOur present study also assessed the status of ferroptosis and EMT in both high- and low-FEIS groups and observed a notable difference. Patients in the high-FEIS group showed a notably subdued ferroptosis status paired with enhanced EMT activity. This pattern was further corroborated by the results of GSEA in GEO and TCGA, which highlighted the substantial enrichment of pathways associated with tumor metastasis and invasion specifically in the high-FEIS group. In summary, the suppressed ferroptosis and active EMT state in the high-FEIS group are important potential factors that contribute to the poor prognosis of patients.\u003c/p\u003e\n\u003cp\u003eWe used ssGSEA and CIBERSORT for assessing immune cell types and quantities in ESCC tissues, evaluating immune pathway activity levels to enhance our comprehension of the immune microenvironment in ESCC. In line with prior studies, our study delineated two distinct subtypes of immune cell infiltration in the context of ssGSEA results[49].\u0026nbsp;The high-FEIS cohort had greater immune cell infiltration than the low-FEIS group, thereby providing a basis for the personalized treatment of different subgroups of patients. Tumor-associated macrophages are known to support tumor growth and metastasis.\u0026nbsp;The ssGSEA analysis showed that M2 macrophages were upregulated in the high-FEIS group, which was strongly linked to tumor advancement and invasion\u0026nbsp;[50]. Chen et al. found that more M2 macrophages could migrate into the tumor microenvironment following FOXO1 induction. By inhibiting antitumor immune responses and promoting angiogenesis, these M2 macrophages further supported the survival and spread of tumor cells, leading to a poor patient prognosis\u0026nbsp;[51]. Immune checkpoint molecules can induce immunosuppressive effects in tumors, thus hindering immune responses against them. In the present study, the expression levels of \u003cem\u003eBTLA\u003c/em\u003e, \u003cem\u003eCTLA4\u003c/em\u003e, \u003cem\u003eICOSLG\u003c/em\u003e, \u003cem\u003ePDCD1LG2\u003c/em\u003e, and \u003cem\u003eTNFRSF9\u003c/em\u003e were upregulated in the high-FEIS group; This finding further validated the role of the biomarkers included in our nomogram model for risk stratification and prognosis assessment.\u003c/p\u003e\n\u003cp\u003ePrevious studies have confirmed that the models based on ferroptosis- or EMT-related genes play important roles in assessing ESCC patient prognosis\u0026nbsp;[52, 53] However, considering the heterogeneity in gene expression and characteristics among different tumor cells and regions in ESCC, predictive methods that rely solely on ferroptosis or EMT-related markers have limitations and may lack high specificity and sensitivity. Compared to the above studies, our study integrated both ferroptosis- and EMT-related genes into a single prognostic model. This combined approach enhanced predictive accuracy compared to models that focused on either pathway alone. Additionally, our model provides the ability to distinguish between different states, including ferroptosis, EMT, and immune suppression, offering a more comprehensive understanding of the tumor microenvironment in ESCC. This contributes valuable insights for personalized treatment strategies, an area that previous models did not address comprehensively. Furthermore, we have included a discussion on the relationship between ACSL3 and VIM expression and ESCC progression, providing a more comprehensive approach to understanding tumor behavior. By comparing our results with these previous studies, the integrated model offers improved predictive power and broader clinical relevance, especially in guiding personalized treatment for ESCC patients.\u003c/p\u003e\n\u003cp\u003eThere are several limitations of this study. First, the data used were obtained from public databases, and the included samples may lack broad representativeness, which limitted the generalizability of the research. Second, the current study is limited by its retrospective nature, which may introduce biases related to sample selection and data collection.\u003c/p\u003e\n\u003cp\u003eIn conclusion, the present study identified a novel prognostic signature based on EMT and ferroptosis for predicting the overall survival of ESCC patients. Additionally, the developed model reflects the immune status of ESCC patients and offers valuable insights for developing personalized treatment strategies for them.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTCGA: https://portal.gdc.cancer.gov/\u003c/p\u003e\n\u003cp\u003eGEO database: https://www.ncbi.nlm.nih.gov/geo/\u003c/p\u003e\n\u003cp\u003eFerrDb website: http://www.zhounan.org/ferrdb\u003c/p\u003e\n\u003cp\u003edbEMT2.0: http://dbemt.bioinfo-minzhao.org/index.html\u003c/p\u003e\n\u003cp\u003eAuthor Contribution\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eZhidong Wang:\u0026nbsp;\u003c/strong\u003eDrafting the manuscript\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCheng Gong:\u003c/strong\u003e Methodology, Software,\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCe Chao:\u0026nbsp;\u003c/strong\u003eResources, Software\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eYoupu Zhang:\u003c/strong\u003e Methodology\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eYongxiang Qian:\u0026nbsp;\u003c/strong\u003eInvestigation, Methodology\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMin Wang:\u0026nbsp;\u003c/strong\u003eSoftware, Funding acquisition\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBin Wang:\u0026nbsp;\u003c/strong\u003ereview and editing of the manuscript\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eYang Liu:\u0026nbsp;\u003c/strong\u003ereview and editing of the manuscript\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by Top Talent of Changzhou \u0026quot;14th Five-Year Plan\u0026quot; High-level Health Personnel Training Project (KY20221388); Major projects of the Changzhou Health Commission (ZD202205); Changzhou Science and Technology Support Plan (Social Development) (CE20205039).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of Interest statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNone\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics and consent statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study received approval from the Clinical Research Ethics Committee of the Third Affiliated Hospital of Soochow University, and informed consent was obtained from all participants. Ethics approval number: 2023(Teaching)CL045-02. Clinical trial number: not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eSung, H., et al., \u003cem\u003eGlobal Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries.\u003c/em\u003e CA Cancer J Clin, 2021. \u003cstrong\u003e71\u003c/strong\u003e(3): p. 209-249.\u003c/li\u003e\n \u003cli\u003eReichenbach, Z.W., et al., \u003cem\u003eClinical and translational advances in esophageal squamous cell carcinoma.\u003c/em\u003e Adv Cancer Res, 2019. \u003cstrong\u003e144\u003c/strong\u003e: p. 95-135.\u003c/li\u003e\n \u003cli\u003eMorgan, E., et al., \u003cem\u003eThe Global Landscape of Esophageal Squamous Cell Carcinoma and Esophageal Adenocarcinoma Incidence and Mortality in 2020 and Projections to 2040: New Estimates From GLOBOCAN 2020.\u003c/em\u003e Gastroenterology, 2022. \u003cstrong\u003e163\u003c/strong\u003e(3): p. 649-658.e2.\u003c/li\u003e\n \u003cli\u003eJiang, X., B.R. Stockwell, and M. Conrad, \u003cem\u003eFerroptosis: mechanisms, biology and role in disease.\u003c/em\u003e Nat Rev Mol Cell Biol, 2021. \u003cstrong\u003e22\u003c/strong\u003e(4): p. 266-282.\u003c/li\u003e\n \u003cli\u003eZhao, L., et al., \u003cem\u003eFerroptosis in cancer and cancer immunotherapy.\u003c/em\u003e Cancer Commun (Lond), 2022. \u003cstrong\u003e42\u003c/strong\u003e(2): p. 88-116.\u003c/li\u003e\n \u003cli\u003eHao, M., et al., \u003cem\u003eFerroptosis regulation by methylation in cancer.\u003c/em\u003e Biochim Biophys Acta Rev Cancer, 2023. \u003cstrong\u003e1878\u003c/strong\u003e(6): p. 188972.\u003c/li\u003e\n \u003cli\u003eJiang, K., et al., \u003cem\u003eSTC2 activates PRMT5 to induce radioresistance through DNA damage repair and ferroptosis pathways in esophageal squamous cell carcinoma.\u003c/em\u003e Redox Biol, 2023. \u003cstrong\u003e60\u003c/strong\u003e: p. 102626.\u003c/li\u003e\n \u003cli\u003eLv, M., et al., \u003cem\u003eCDK7-YAP-LDHD axis promotes D-lactate elimination and ferroptosis defense to support cancer stem cell-like properties.\u003c/em\u003e Signal Transduct Target Ther, 2023. \u003cstrong\u003e8\u003c/strong\u003e(1): p. 302.\u003c/li\u003e\n \u003cli\u003eShishido, Y., et al., \u003cem\u003eAntitumor Effect of 5-Aminolevulinic Acid Through Ferroptosis in Esophageal Squamous Cell Carcinoma.\u003c/em\u003e Ann Surg Oncol, 2021. \u003cstrong\u003e28\u003c/strong\u003e(7): p. 3996-4006.\u003c/li\u003e\n \u003cli\u003eDongre, A. and R.A. Weinberg, \u003cem\u003eNew insights into the mechanisms of epithelial-mesenchymal transition and implications for cancer.\u003c/em\u003e Nat Rev Mol Cell Biol, 2019. \u003cstrong\u003e20\u003c/strong\u003e(2): p. 69-84.\u003c/li\u003e\n \u003cli\u003eDebnath, P., et al., \u003cem\u003eEpithelial-mesenchymal transition and its transcription factors.\u003c/em\u003e Biosci Rep, 2022. \u003cstrong\u003e42\u003c/strong\u003e(1).\u003c/li\u003e\n \u003cli\u003eMittal, V., \u003cem\u003eEpithelial Mesenchymal Transition in Tumor Metastasis.\u003c/em\u003e Annu Rev Pathol, 2018. \u003cstrong\u003e13\u003c/strong\u003e: p. 395-412.\u003c/li\u003e\n \u003cli\u003eMortezaee, K., J. Majidpoor, and E. Kharazinejad, \u003cem\u003eEpithelial-mesenchymal transition in cancer stemness and heterogeneity: updated.\u003c/em\u003e Med Oncol, 2022. \u003cstrong\u003e39\u003c/strong\u003e(12): p. 193.\u003c/li\u003e\n \u003cli\u003eLiu, X., et al., \u003cem\u003eEMT and Cancer Cell Stemness Associated With Chemotherapeutic Resistance in Esophageal Cancer.\u003c/em\u003e Front Oncol, 2021. \u003cstrong\u003e11\u003c/strong\u003e: p. 672222.\u003c/li\u003e\n \u003cli\u003eZhao, L., et al., \u003cem\u003eIntegrative network biology analysis identifies miR-508-3p as the determinant for the mesenchymal identity and a strong prognostic biomarker of ovarian cancer.\u003c/em\u003e Oncogene, 2019. \u003cstrong\u003e38\u003c/strong\u003e(13): p. 2305-2319.\u003c/li\u003e\n \u003cli\u003eCong, Z., et al., \u003cem\u003eBTB domain and CNC homolog 1 promotes glioma invasion mainly through regulating extracellular matrix and increases ferroptosis sensitivity.\u003c/em\u003e Biochim Biophys Acta Mol Basis Dis, 2022. \u003cstrong\u003e1868\u003c/strong\u003e(12): p. 166554.\u003c/li\u003e\n \u003cli\u003eLiu, L., et al., \u003cem\u003eFerroptosis: Mechanism and connections with cutaneous diseases.\u003c/em\u003e Front Cell Dev Biol, 2022. \u003cstrong\u003e10\u003c/strong\u003e: p. 1079548.\u003c/li\u003e\n \u003cli\u003eGuan, D., et al., \u003cem\u003eThe DpdtbA induced EMT inhibition in gastric cancer cell lines was through ferritinophagy-mediated activation of p53 and PHD2/hif-1\u0026alpha; pathway.\u003c/em\u003e J Inorg Biochem, 2021. \u003cstrong\u003e218\u003c/strong\u003e: p. 111413.\u003c/li\u003e\n \u003cli\u003eYao, J., et al., \u003cem\u003eSingle-Cell RNA-Seq Reveals the Promoting Role of Ferroptosis Tendency During Lung Adenocarcinoma EMT Progression.\u003c/em\u003e Front Cell Dev Biol, 2021. \u003cstrong\u003e9\u003c/strong\u003e: p. 822315.\u003c/li\u003e\n \u003cli\u003eZhang, W., et al., \u003cem\u003eResveratrol inhibits ferroptosis and decelerates heart failure progression via Sirt1/p53 pathway activation.\u003c/em\u003e J Cell Mol Med, 2023. \u003cstrong\u003e27\u003c/strong\u003e(20): p. 3075-3089.\u003c/li\u003e\n \u003cli\u003eBao, Z., et al., \u003cem\u003eMEF2C silencing downregulates NF2 and E-cadherin and enhances Erastin-induced ferroptosis in meningioma.\u003c/em\u003e Neuro Oncol, 2021. \u003cstrong\u003e23\u003c/strong\u003e(12): p. 2014-2027.\u003c/li\u003e\n \u003cli\u003eKlasson, T.D., et al., \u003cem\u003eACSL3 regulates lipid droplet biogenesis and ferroptosis sensitivity in clear cell renal cell carcinoma.\u003c/em\u003e Cancer Metab, 2022. \u003cstrong\u003e10\u003c/strong\u003e(1): p. 14.\u003c/li\u003e\n \u003cli\u003eQin, Z., et al., \u003cem\u003eDesign and synthesis of isothiocyanate-containing hybrid androgen receptor (AR) antagonist to downregulate AR and induce ferroptosis in GSH-Deficient prostate cancer cells.\u003c/em\u003e Chem Biol Drug Des, 2021. \u003cstrong\u003e97\u003c/strong\u003e(5): p. 1059-1078.\u003c/li\u003e\n \u003cli\u003eSong, X., et al., \u003cem\u003ePDK4 dictates metabolic resistance to ferroptosis by suppressing pyruvate oxidation and fatty acid synthesis.\u003c/em\u003e Cell Rep, 2021. \u003cstrong\u003e34\u003c/strong\u003e(8): p. 108767.\u003c/li\u003e\n \u003cli\u003eZhu, J., et al., \u003cem\u003eThe deubiquitinase USP11 ameliorates intervertebral disc degeneration by regulating oxidative stress-induced ferroptosis via deubiquitinating and stabilizing Sirt3.\u003c/em\u003e Redox Biol, 2023. \u003cstrong\u003e62\u003c/strong\u003e: p. 102707.\u003c/li\u003e\n \u003cli\u003eQiao, J., et al., \u003cem\u003eNR5A2 synergizes with NCOA3 to induce breast cancer resistance to BET inhibitor by upregulating NRF2 to attenuate ferroptosis.\u003c/em\u003e Biochem Biophys Res Commun, 2020. \u003cstrong\u003e530\u003c/strong\u003e(2): p. 402-409.\u003c/li\u003e\n \u003cli\u003eHao, L., et al., \u003cem\u003eSLC40A1 Mediates Ferroptosis and Cognitive Dysfunction in Type 1 Diabetes.\u003c/em\u003e Neuroscience, 2021. \u003cstrong\u003e463\u003c/strong\u003e: p. 216-226.\u003c/li\u003e\n \u003cli\u003eWang, H., et al., \u003cem\u003eSilencing of PTPN18 Induced Ferroptosis in Endometrial Cancer Cells Through p-P38-Mediated GPX4/xCT Down-Regulation.\u003c/em\u003e Cancer Manag Res, 2021. \u003cstrong\u003e13\u003c/strong\u003e: p. 1757-1765.\u003c/li\u003e\n \u003cli\u003eWang, Z., et al., \u003cem\u003eProtein and metabolic profiles of tyrosine kinase inhibitors co-resistant liver cancer cells.\u003c/em\u003e Front Pharmacol, 2024. \u003cstrong\u003e15\u003c/strong\u003e: p. 1394241.\u003c/li\u003e\n \u003cli\u003eLi, M., et al., \u003cem\u003eCurcumin inhibits the invasion and metastasis of triple negative breast cancer via Hedgehog/Gli1 signaling pathway.\u003c/em\u003e J Ethnopharmacol, 2022. \u003cstrong\u003e283\u003c/strong\u003e: p. 114689.\u003c/li\u003e\n \u003cli\u003eZhang, W., W. Liu, and X. Hu, \u003cem\u003eRobinin inhibits pancreatic cancer cell proliferation, EMT and inflammation via regulating TLR2-PI3k-AKT signaling pathway.\u003c/em\u003e Cancer Cell Int, 2023. \u003cstrong\u003e23\u003c/strong\u003e(1): p. 328.\u003c/li\u003e\n \u003cli\u003eGupta, N., et al., \u003cem\u003eNotch3 induces epithelial-mesenchymal transition and attenuates carboplatin-induced apoptosis in ovarian cancer cells.\u003c/em\u003e Gynecol Oncol, 2013. \u003cstrong\u003e130\u003c/strong\u003e(1): p. 200-6.\u003c/li\u003e\n \u003cli\u003eArrindell, J. and B. Desnues, \u003cem\u003eVimentin: from a cytoskeletal protein to a critical modulator of immune response and a target for infection.\u003c/em\u003e Front Immunol, 2023. \u003cstrong\u003e14\u003c/strong\u003e: p. 1224352.\u003c/li\u003e\n \u003cli\u003eIvaska, J., et al., \u003cem\u003eNovel functions of vimentin in cell adhesion, migration, and signaling.\u003c/em\u003e Exp Cell Res, 2007. \u003cstrong\u003e313\u003c/strong\u003e(10): p. 2050-62.\u003c/li\u003e\n \u003cli\u003eLi, C. and Y.Q. Ma, \u003cem\u003ePrognostic significance of sex determining region Y-box 2, E-cadherin, and vimentin in esophageal squamous cell carcinoma.\u003c/em\u003e World J Clin Cases, 2022. \u003cstrong\u003e10\u003c/strong\u003e(27): p. 9657-9669.\u003c/li\u003e\n \u003cli\u003eQuan, J., A.M. Bode, and X. Luo, \u003cem\u003eACSL family: The regulatory mechanisms and therapeutic implications in cancer.\u003c/em\u003e Eur J Pharmacol, 2021. \u003cstrong\u003e909\u003c/strong\u003e: p. 174397.\u003c/li\u003e\n \u003cli\u003eMagtanong, L., et al., \u003cem\u003eExogenous Monounsaturated Fatty Acids Promote a Ferroptosis-Resistant Cell State.\u003c/em\u003e Cell Chem Biol, 2019. \u003cstrong\u003e26\u003c/strong\u003e(3): p. 420-432.e9.\u003c/li\u003e\n \u003cli\u003eSaliakoura, M., et al., \u003cem\u003eThe ACSL3-LPIAT1 signaling drives prostaglandin synthesis in non-small cell lung cancer.\u003c/em\u003e Oncogene, 2020. \u003cstrong\u003e39\u003c/strong\u003e(14): p. 2948-2960.\u003c/li\u003e\n \u003cli\u003eYin, L., et al., \u003cem\u003eHOOK1 Inhibits the Progression of Renal Cell Carcinoma via TGF-\u0026beta; and TNFSF13B/VEGF-A Axis.\u003c/em\u003e Adv Sci (Weinh), 2023. \u003cstrong\u003e10\u003c/strong\u003e(17): p. e2206955.\u003c/li\u003e\n \u003cli\u003eChen, K., et al., \u003cem\u003eMAP4K4 promotes ovarian cancer metastasis through diminishing ADAM10-dependent N-cadherin cleavage.\u003c/em\u003e Oncogene, 2023. \u003cstrong\u003e42\u003c/strong\u003e(18): p. 1438-1452.\u003c/li\u003e\n \u003cli\u003eDelpire, E., \u003cem\u003eThe mammalian family of sterile 20p-like protein kinases.\u003c/em\u003e Pflugers Arch, 2009. \u003cstrong\u003e458\u003c/strong\u003e(5): p. 953-67.\u003c/li\u003e\n \u003cli\u003eSun, R., et al., \u003cem\u003eAnnexin10 promotes extrahepatic cholangiocarcinoma metastasis by facilitating EMT via PLA2G4A/PGE2/STAT3 pathway.\u003c/em\u003e EBioMedicine, 2019. \u003cstrong\u003e47\u003c/strong\u003e: p. 142-155.\u003c/li\u003e\n \u003cli\u003eKaufmann, R., et al., \u003cem\u003eThrombin-mediated hepatocellular carcinoma cell migration: cooperative action via proteinase-activated receptors 1 and 4.\u003c/em\u003e J Cell Physiol, 2007. \u003cstrong\u003e211\u003c/strong\u003e(3): p. 699-707.\u003c/li\u003e\n \u003cli\u003eMou, Y., et al., \u003cem\u003eFerroptosis, a new form of cell death: opportunities and challenges in cancer.\u003c/em\u003e J Hematol Oncol, 2019. \u003cstrong\u003e12\u003c/strong\u003e(1): p. 34.\u003c/li\u003e\n \u003cli\u003eGabbs, M., et al., \u003cem\u003eAdvances in Our Understanding of Oxylipins Derived from Dietary PUFAs.\u003c/em\u003e Adv Nutr, 2015. \u003cstrong\u003e6\u003c/strong\u003e(5): p. 513-40.\u003c/li\u003e\n \u003cli\u003eYang, X., et al., \u003cem\u003emiR-18a promotes glioblastoma development by down-regulating ALOXE3-mediated ferroptotic and anti-migration activities.\u003c/em\u003e Oncogenesis, 2021. \u003cstrong\u003e10\u003c/strong\u003e(2): p. 15.\u003c/li\u003e\n \u003cli\u003eEndo, M., \u003cem\u003eThe Roles of ANGPTL Families in Cancer Progression.\u003c/em\u003e J uoeh, 2019. \u003cstrong\u003e41\u003c/strong\u003e(3): p. 317-325.\u003c/li\u003e\n \u003cli\u003eYang, W.H., et al., \u003cem\u003eA TAZ-ANGPTL4-NOX2 Axis Regulates Ferroptotic Cell Death and Chemoresistance in Epithelial Ovarian Cancer.\u003c/em\u003e Mol Cancer Res, 2020. \u003cstrong\u003e18\u003c/strong\u003e(1): p. 79-90.\u003c/li\u003e\n \u003cli\u003eZhuang, W., et al., \u003cem\u003eAn immunogenomic signature for molecular classification in hepatocellular carcinoma.\u003c/em\u003e Mol Ther Nucleic Acids, 2021. \u003cstrong\u003e25\u003c/strong\u003e: p. 105-115.\u003c/li\u003e\n \u003cli\u003eYunna, C., et al., \u003cem\u003eMacrophage M1/M2 polarization.\u003c/em\u003e Eur J Pharmacol, 2020. \u003cstrong\u003e877\u003c/strong\u003e: p. 173090.\u003c/li\u003e\n \u003cli\u003eWang, Y., et al., \u003cem\u003eFOXO1 promotes tumor progression by increased M2 macrophage infiltration in esophageal squamous cell carcinoma.\u003c/em\u003e Theranostics, 2020. \u003cstrong\u003e10\u003c/strong\u003e(25): p. 11535-11548.\u003c/li\u003e\n \u003cli\u003eChen, L., et al., \u003cem\u003eAberrant epithelial cell interaction promotes esophageal squamous-cell carcinoma development and progression.\u003c/em\u003e Signal Transduct Target Ther, 2023. \u003cstrong\u003e8\u003c/strong\u003e(1): p. 453.\u003c/li\u003e\n \u003cli\u003eSong, J., et al., \u003cem\u003eA Novel Ferroptosis-Related Biomarker Signature to Predict Overall Survival of Esophageal Squamous Cell Carcinoma.\u003c/em\u003e Front Mol Biosci, 2021. \u003cstrong\u003e8\u003c/strong\u003e: p. 675193.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"ferroptosis, epithelial-mesenchymal transition, GEO, TCGA, bioinformatics analysis, esophageal squamous cell carcinoma","lastPublishedDoi":"10.21203/rs.3.rs-6456491/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6456491/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e: Limited research has been conducted on the interaction between ferroptosis and epithelial-mesenchymal transition (EMT) and their combined effect on esophageal squamous cell carcinoma (ESCC) patient prognosis. The present study aimed to develop a prognostic model based on the impact of ferroptosis and EMT on ESCC prognosis for clinical application.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e: Gene expression levels and clinical data of ESCC patients were obtained from the GSE53625 dataset in the gene expression omnibus (GEO) database, and the data from the cancer genome atlas (TCGA) were obtained as a validation set. By combining the results of cox regression analysis and least absolute shrinkage and selection operator regression (LASSO) analysis, we selected nine genes associated with prognosis, which were then used to construct a prognostic model. Immune cell infiltration was evaluated using CIBERSORT and single-sample Gene Set Enrichment Analysis methods.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e: Nine key genes were screened to construct ferroptosis and EMT integrated score (FEIS). Compared to the low-FEIS group, the high-FEIS group demonstrated shorter overall survival period. The immune infiltration analysis showed an increase in immune cell infiltration and elevated expression levels of immune checkpoint molecules in the high-FEIS group. A nomogram was constructed to accurately predict patient prognosis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e: Our study introduced a novel prognostic tool that integrates ferroptosis -and EMT-related biomarker, and offered valuable insights for developing personalized treatment strategies for ESCC patients.\u003c/p\u003e","manuscriptTitle":"A dual EMT-ferroptosis gene signature predicts survival and immune infiltration in esophageal squamous cell carcinoma","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-09 11:14:43","doi":"10.21203/rs.3.rs-6456491/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"31f0634c-e586-470b-9912-ca14f4801255","owner":[],"postedDate":"May 9th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-06-10T10:23:58+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-09 11:14:43","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6456491","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6456491","identity":"rs-6456491","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}