Identifying High-Risk Features Associated with Limited Overall Survival in Esophageal Cancer Patients with Vascular Invasion Using Machine Learning

Identifying High-Risk Features Associated with Limited Overall Survival in Esophageal Cancer Patients with Vascular Invasion Using Machine Learning | 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 Article Identifying High-Risk Features Associated with Limited Overall Survival in Esophageal Cancer Patients with Vascular Invasion Using Machine Learning Yue Zhao, Jun-jie Liu, zhan zhang, Zhen-yi Li, Yi-jun Ma, Si-jie Zhang, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7151065/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Background Vascular invasion is a critical pathological feature associated with poor prognosis in esophageal cancer, yet individualized risk prediction in this subgroup remains limited. Methods We retrospectively analyzed clinical data from 318 esophageal cancer patients with confirmed vascular invasion who underwent surgical resection at Shandong University Qilu Hospital between January 2019 and December 2022. Eight machine learning models were constructed using clinical, pathological, and laboratory features. The Gradient Boosting Machine (GBM) model was selected based on superior performance in discrimination, calibration, and decision curve analysis. Internal validation and survival stratification were conducted to assess robustness and clinical utility. Results Eight variables were identified as independent prognostic factors: nerve invasion, invasion of the fibrous outer membrane, T stage, N stage, BMI, white blood cell count, squamous cell carcinoma antigen level, and total number of lymph nodes dissected. The GBM model achieved the highest time-dependent AUCs (1-year: 0.987, 2-year: 0.971, 3-year: 0.976) and demonstrated consistent calibration and net clinical benefit. Survival analysis based on GBM risk scores revealed significant stratification between risk groups (p < 0.001). Conclusion Our GBM-based model provides accurate and interpretable prognostic predictions for esophageal cancer patients with vascular invasion. It offers valuable guidance for individualized clinical decision-making and warrants further validation in multicenter prospective cohorts. Biological sciences/Cancer Health sciences/Oncology esophageal cancer vascular invasion prognostic model machine learning survival prediction risk stratification Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Esophageal cancer (EC) remains a significant global health burden, ranking as the seventh most commonly diagnosed cancer and the sixth leading cause of cancer-related mortality worldwide( 1 , 2 ). According to GLOBOCAN 2020 data, an estimated 604,000 new EC cases and 544,000 related deaths were reported globally in that year( 3 ). Despite advancements in surgical techniques, chemotherapy, radiotherapy, and immunotherapy, the overall 5-year survival rate remains below 20%, largely due to the late-stage diagnosis and high recurrence rates( 4 ). Vascular invasion (VI), which includes both blood vessel invasion and lymphatic vessel invasion, is a critical pathological feature that significantly impacts disease progression( 5 ). Numerous studies have demonstrated that the presence of vascular invasion is associated with higher recurrence rates, increased distant metastasis, and reduced survival outcomes, particularly progression-free survival (PFS) and overall survival (OS)( 6 , 7 ). Accurate identification of high-risk factors for shortened OS in patients with vascular invasion is essential for optimizing treatment strategies. Traditionally, prognosis is evaluated based on clinical staging systems such as the TNM classification, which considers tumor size, lymph node involvement, and distant metastasis( 8 ). However, vascular invasion is not explicitly included as a prognostic factor in the TNM system, despite its well-documented impact on recurrence and survival. This limitation is particularly relevant in early-stage esophageal cancer, where treatment decisions are often conservative when guided solely by TNM staging. In patients with positive vascular invasion, the biological behavior of the tumor may be far more aggressive than the staging suggests( 9 ). Vascular invasion is increasingly recognized as a marker of occult metastatic potential and is frequently associated with poorer clinical outcomes( 10 ). Xie et al. reported markedly reduced 1-, 3-, and 5-year survival rates in VI-positive compared to VI-negative patients (93.4% vs. 66.7%, 53.8% vs. 18.8%, and 48.1% vs. 15.6%, respectively)( 11 ). Even among patients classified as early stage, the presence of vascular invasion was independently associated with significantly worse overall survival( 12 ). These findings suggest that applying standard treatment protocols to VI-positive patients may lead to undertreatment in a subgroup with high risk of progression. Therefore, further risk stratification through the identification of additional prognostic indicators that contribute to poor prognosis in this patient population is warranted. Such efforts could inform decisions regarding the intensity of adjuvant therapy, frequency of follow-up, and need for more aggressive intervention, ultimately facilitating a more individualized and risk-adapted treatment approach. In recent years, machine learning (ML) has emerged as a powerful tool in oncologic research, offering improved predictive accuracy by integrating complex clinical, radiological, and pathological data( 13 , 14 ). ML models have been successfully applied to predict tumor recurrence, treatment response, and survival outcomes in various cancers, including lung, gastric, and colorectal cancers( 15 , 16 ). In esophageal cancer, ML-based approaches have been utilized for tumor classification, lymph node metastasis prediction, and response assessment to neoadjuvant therapy( 17 ). However, limited studies have focused on leveraging ML for predicting high-risk factors associated with shortened OS in esophageal cancer patients with vascular invasion. This study aims to develop and validate a machine learning-based model to identify high-risk features associated with limited OS in esophageal cancer patients with vascular invasion. By incorporating demographic, clinicopathologic, and radiological features, this model seeks to enhance risk stratification and improve preoperative decision-making. Identifying these prognostic factors can help refine treatment strategies, potentially guiding more aggressive surgical resections, adjuvant therapy recommendations, and closer postoperative surveillance for high-risk individuals. Patients and methods Patients’ selection we systematically analyzed clinical data from a total of 354 patients who underwent surgical resection for esophageal cancer at Shandong University Qilu Hospital over a period spanning from January 2019 to December 2022. Patients were enrolled in this study based on the following criteria: ( 1 ) Histologically confirmed diagnosis of esophageal squamous cell carcinoma or adenocarcinoma with positive vascular invasion confirmed by postoperative pathological examination; ( 2 ) Underwent curative-intent surgical resection, including esophagectomy combined with regional lymph node dissection; ( 3 ) Availability of complete clinicopathological data, including preoperative imaging, surgical records, and pathological reports; ( 4 ) Adequate follow-up data for progression-free survival analysis; ( 5 ) No prior history of other malignancies that could affect vascular invasion status. Patients were excluded if they had a non-esophageal primary tumor, lacked evidence of vascular invasion, had incomplete clinical or pathological data, underwent non-curative surgery, lacked adequate follow-up information for overall survival analysis, or experienced perioperative mortality unrelated to tumor progression. Finally, we excluded 36 patients due to loss of follow-up. Based on strict adherence to the inclusion and exclusion criteria, a total of 318 patients were ultimately included in this study. This study received approval from the Medical Ethics Committee of Shandong University Qilu Hospital (approval number: KYLL-202008-023-1). Given the retrospective nature of this study utilizing previously collected clinical records, the institutional ethics committee granted a waiver of informed consent, and rigorous measures were implemented to protect patient confidentiality throughout the research process. Data collection We collected a comprehensive range of clinical and pathological variables for all eligible patients. The dataset included: ( 1 ) demographic characteristics such as sex, age, weight, body mass index (BMI), smoking and alcohol history, and comorbidities; ( 2 ) tumor-related features, including tumor location, length and adventitial invasion, as measured by preoperative imaging or endoscopy; ( 3 ) preoperative laboratory indicators including hematologic parameters, liver and renal function, glucose and lipid metabolism, and tumor markers; ( 4 ) treatment-related information including surgical approach, type of resection, and perioperative therapy; ( 5 ) Pathological features such as histological type, tumor differentiation grade, T stage, N stage, and nerve invasion; ( 6 ) follow-up data including survival status, survival time. To ensure data completeness and minimize potential bias, we applied a two-tiered approach. Variables with more than 30% missing data were excluded from further analysis to maintain data reliability. For variables with less than 30% missingness, missing values in continuous variables were imputed using the mean of the respective variable, while those in categorical variables were filled using the mode. Tumor-related features Tumor-related characteristics included tumor location, tumor length and Invasion of fibrous outer membrane. Tumor location was categorized as upper thoracic, middle thoracic, or lower thoracic esophagus, based on the anatomical segmentation defined by the 8th edition of the American Joint Committee on Cancer (AJCC) TNM staging system( 18 ). All patients underwent upper gastrointestinal radiography and contrast-enhanced computed tomography (CT) scans of the chest and upper abdomen within two weeks prior to surgery. The primary tumor site and longitudinal extent were primarily determined through gastrointestinal endoscopy. In cases where endoscopic documentation was incomplete or lacked specific measurements, tumor length was estimated using a combination of upper gastrointestinal contrast studies and contrast-enhanced-CT imaging. Invasion of the fibrous outer membrane was independently reviewed by two board-certified radiologists, each with over ten years of experience in thoracic oncology imaging. When imaging findings could not provide a definitive assessment, intraoperative observations were additionally considered for final documentation. Any discrepancies in interpretation were resolved by consensus discussion. Pathological diagnosis The assessment of vascular invasion was performed on routinely processed postoperative specimens. All resected tissues were fixed in 10% neutral-buffered formalin, embedded in paraffin, and sectioned at 4-µm thickness. Hematoxylin and eosin (HE) staining was conducted in accordance with standardized histopathological protocols. Two experienced pathologists, blinded to the clinical characteristics and outcomes, independently evaluated each slide using light microscopy. Elastic or immunohistochemical staining was additionally employed when morphological assessment alone was insufficient to definitively identify vascular structures. Vascular invasion was defined as the unequivocal presence of tumor cells within the lumen of blood vessels or lymphatic channels, irrespective of the presence of an endothelial lining( 5 ). Statistical analysis Statistical analyses were conducted using R software (version 4.3.3). The analysis utilized several R packages, including randomForestSRC, CoxBoost, gbm, xgboost, glmnet, superpc, plsRcox and others. The 318 patients were randomized into training cohorts(TC) and validation cohorts(VC) in a 7:3 ratio using a random number seed. Continuous variables were summarized as mean ± standard deviation (SD) when normally distributed. Comparisons between groups were carried out using the Student’s t-test or Mann–Whitney U test, depending on data distribution. Categorical variables were described as counts and percentages, with group differences assessed using the chi-square test or Fisher’s exact test, as appropriate. OS was defined as the time from surgery to death from any cause or last follow-up. Kaplan–Meier survival curves were constructed to estimate OS, and statistical differences between subgroups were evaluated using the log-rank test. To explore prognostic factors, univariate Cox proportional hazards regression was first conducted. Variables demonstrating a P-value < 0.05 in univariate analysis were subsequently entered into a multivariate Cox model to identify independent predictors of limited OS. A two-sided P-value < 0.05 was considered to indicate statistical significance across all tests. Building Machine Learning Models Based on feature selection and clinical relevance, eight key variables were incorporated into the modeling process, and eight machine learning algorithms were subsequently applied to analyze high-risk features and predict overall survival (OS) in patients with esophageal cancer and vascular invasion. These algorithms included Random Survival Forest (RSF), Gradient Boosting Machine (GBM), Least Absolute Shrinkage and Selection Operator Cox Regression (LASSO-Cox), Cox Proportional Hazards Model with Boosting (CoxBoost), Survival Support Vector Machine (survivalsvm), Extreme Gradient Boosting (XGBoost), Supervised Principal Components (SuperPC), Partial Least Squares Regression for Cox (plsRcox). Model performance was comprehensively evaluated using multiple metrics, including time-dependent area under the curve (Time-AUC), concordance index (C-index), receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis (DCA), across both training and validation cohorts. Time-AUC curves were generated to assess the dynamic predictive ability of each model at 1-, 2-, and 3-year time points, providing insights into their temporal discriminative power. Calibration plots were used to assess the agreement between predicted and observed outcomes, while DCA evaluated the net clinical benefit of each model across a range of threshold probabilities. Results Baseline Features of Patients A total of 318 esophageal cancer patients with vascular invasion from Qilu Hospital of Shandong University were included in this study, with 223 patients assigned to the training cohort and 95 to the validation cohort (Figure 1) . Baseline characteristics are summarized in Table 1 . Each cohort was further stratified into survival subgroups to facilitate comparison. In both cohorts, the mean age was approximately 63 years, and the male predominance was consistent (TC: 81.8%–86.7%, VC: 86.0%–86.7%). Key demographic variables, including weight, BMI, smoking and alcohol history, hypertension, and diabetes, were comparable between alive and deceased groups within each cohort ( P > 0.05). Notable tumor-related differences were observed in both cohorts. Deceased patients had significantly larger tumors and a higher incidence of adventitial invasion compared to survivors (P < 0.01 in both TC and VC). Among laboratory indicators, higher white blood cell count (WBC), monocyte count, platelet count (PLT), serum sialic acid (SA), squamous cell carcinoma antigen (SCC), and serum ferritin (SF) levels were observed in deceased patients compared to survivors, with several of these variables showing significant differences. Albumin levels were significantly higher in survivors in the validation cohort (P = 0.041). Moreover, advanced T and N stage, as well as higher rates of nerve invasion (NI), were significantly associated with mortality in both cohorts ( P < 0.05). To balance the two cohorts, baseline characteristics were also compared between the training and validation groups in table1 . Except for differences in triglyceride levels, serum sialic acid, and the total number of dissected lymph nodes, all other variables showed no statistically significant differences between the two cohorts. Consequently, these findings support the use of the validation cohort for further assessment of model performance. Risk Factor Selection and Model Development To further identify prognostic variables associated with limited OS, we performed univariate and multivariate Cox proportional hazards regression analyses. The results are summarized in table 2 . In the univariate Cox regression analysis, eleven variables were found to be significantly associated with overall survival (OS), including BMI, tumor size, invasion of the fibrous outer membrane, WBC count, SA, SCC, SF, T stage, N stage, NI, and the total number of dissected lymph nodes (P < 0.05 for all). Variables with statistical significance in univariate analysis were subsequently included in the multivariate Cox regression model. The multivariate analysis identified the following eight variables as independent prognostic factors for limited OS: lower BMI (HR = 0.93, 95% CI: 0.87–0.99, P = 0.019), invasion of the fibrous outer membrane (HR = 1.74, 95% CI: 1.14–2.65, P = 0.011), elevated WBC count (HR = 1.19, 95% CI: 1.06–1.34, P = 0.002), higher SCC levels (HR = 1.12, 95% CI: 1.04–1.20, P = 0.002), advanced T stage (T3: HR = 4.65, 95% CI: 1.09–19.87, P = 0.038; T4: HR = 6.31, 95% CI: 1.27–31.28, P = 0.024), advanced N stage (N1: HR = 1.77, 95% CI: 1.00–3.11, P = 0.049; N2: HR = 1.98, 95% CI: 1.08–3.61, P = 0.027; N3: HR = 2.43, 95% CI: 1.25–4.73, P = 0.009), presence of NI (HR = 1.54, 95% CI: 1.01–2.34, P = 0.043), and total number of lymph nodes (HR = 1.03, 95% CI: 1.00–1.05, P = 0.022). Using these selected variables, eight machine learning algorithms were developed to construct prognostic models: RSF, GBM, LASSO-Cox, CoxBoost, SurvivalSVM, XGBoost, SuperPC, and PLSR-Cox. Model performance was further systematically evaluated in both the training and validation cohorts to confirm predictive accuracy, model stability, and potential clinical relevance in survival risk stratification. Evaluation of Prognostic Models Based on Eight Machine Learning The predictive performance of the eight machine learning–based prognostic models was evaluated using time-dependent ROC curves at the 1-, 2-, and 3-year time points, as well as continuously measured area under the curve (AUC) values over the entire follow-up period. As shown in Figure 2 , at 1 year, GBM achieved the highest AUC (0.987), followed closely by RSF (0.946) and XGBoost (0.945). At 2 and 3 years, GBM (AUC = 0.971 and 0.976, respectively) and RSF (AUC = 0.968 and 0.970) remained top performers. Specifically, GBM and RSF consistently demonstrated superior discriminative performance across all time points, with AUCs exceeding 0.94 at each interval. In contrast, models such as SurvivalSVM and SuperPC exhibited relatively lower discriminative ability, with AUCs ranging from 0.523 to 0.712 over the three time points. LASSO-Cox, CoxBoost, and PLSR-Cox models showed moderate performance, with AUC values generally between 0.75 and 0.85. To further assess the agreement between predicted and observed survival probabilities, calibration curves were plotted for each model at the 1-, 2-, and 3-year time points. As shown in the calibration plots (Figure 3a and Additional file 1) , most models demonstrated acceptable alignment with the 45-degree reference line, indicating reasonable calibration performance. Among them, the GBM model exhibited the most consistent and closest fit to the ideal line across all three time points, reflecting excellent agreement between predicted and actual overall survival probabilities. Additionally, The RSF models also demonstrated good calibration, particularly at the 1- and 2-year marks, though slight deviations were noted at 3 years (Additional file 1b) . Moreover, to evaluate the potential clinical utility of each prognostic model, DCA was also conducted at the 1-, 2-, and 3-year time points (Figure3) . DCA estimates the net benefit of a prediction model across a range of threshold probabilities, thereby reflecting its value in assisting clinical decision-making. As shown in the DCA curves, the GBM model consistently yielded the highest net benefit across most threshold probabilities at all three time points. Although the RSF model exhibited relatively lower net benefit at the 1-year mark, its performance improved at the 2- and 3-year time points, showing moderate clinical utility in longer-term prognostic prediction. Taken together, considering its superior discrimination, excellent calibration, and the highest net clinical benefit across all time points, the GBM model was selected as the optimal prognostic model for overall survival prediction in this cohort. Internal Validation and Survival Stratification To further evaluate the generalizability of the GBM model, internal validation was performed using the independent test cohort (Figure4) . The model exhibited strong discriminatory capability, as evidenced by AUCs of 0.786, 0.774, and 0.750 at 1, 2, and 3 years, respectively, with consistent performance over time. Calibration plots also revealed a strong agreement between predicted and observed survival probabilities at 1, 2, and 3 years. These findings further support the robust performance and clinical applicability of the GBM model in prognostic risk estimation. Furthermore, Kaplan–Meier survival analysis was performed to evaluate the stratification ability of the GBM model (Figure5) . Based on the predicted risk scores, patients were divided into high-risk and low-risk groups using the median score as the cutoff. As shown in the survival curves, the high-risk group exhibited significantly worse OS compared to the low-risk group in both the training cohort (log-rank test, P < 0.0001) and the validation cohort ( P = 0.0004). This clear separation between the two groups demonstrates the strong prognostic discriminatory power of the GBM model and its potential utility in individualized risk stratification. Discussion In recent years, vascular invasion has gained increasing attention as a crucial pathological feature associated with poor prognosis and early recurrence in esophageal cancer. Defined as the presence of tumor cells within blood or lymphatic vessels, vascular invasion represents a direct route for tumor dissemination and metastasis. Multiple studies have demonstrated that the presence of vascular invasion correlates strongly with adverse clinical outcomes, including decreased PFS and OS (19, 20). As such, patients with vascular invasion represent a clinically distinct subgroup that may benefit from intensified surveillance, enhanced postoperative monitoring, or escalated adjuvant treatment strategies. Despite its well-recognized prognostic significance, vascular invasion has not yet been incorporated into standard staging systems such as the TNM classification, which may lead to an underestimation of recurrence risk in affected patients. Our study aims to leverage machine learning techniques to identify high-risk factors associated with limited overall survival in esophageal cancer patients with vascular invasion, thereby enhancing precision in clinical decision-making and improving outcomes for this vulnerable patient population. In this study, we developed and validated multiple machine learning–based prognostic models to predict overall survival in esophageal cancer patients with vascular invasion. By incorporating diverse clinical, pathological, imaging, and laboratory features, we successfully developed machine learning models to predict prognosis in esophageal cancer patients with vascular invasion. Compared to traditional survival modeling methods, such as Cox regression and traditional nomogram models, machine learning approaches allow for the integration of nonlinear interactions and high-dimensional feature spaces. In particular, we employed eight distinct algorithms spanning tree-based models (e.g., RSF, GBM, XGBoost), regularized regression models (e.g., LASSO-Cox, CoxBoost), and other survival learning frameworks (e.g., plsRcox, superPC, survivalSVM), enabling a robust comparative evaluation. Among all models, the GBM consistently outperformed others in both discrimination and calibration metrics, with the highest time-dependent AUCs at 1-, 2-, and 3-year follow-up points and strong net clinical benefit across decision curve analyses. GBM is a powerful ensemble learning algorithm that builds models sequentially, where each iteration focuses on minimizing the prediction error of the previous model(21). The GBM model’s superior performance can be attributed to its ability to iteratively refine predictions through boosting, effectively capturing subtle feature interactions and mitigating overfitting via embedded regularization technique(22). Survival curve stratification based on GBM risk scores also demonstrated clear prognostic separation, underscoring its clinical utility in individualized risk assessment. Through univariate and multivariate Cox regression analyses, eight independent prognostic factors were identified—including fibrous outer membrane invasion, nerve invasion, T stage, N stage, BMI, WBC count, SCC level, and total number of lymph nodes dissected—for inclusion in the final prognostic model. Among esophageal cancer patients with vascular invasion, both invasion of the fibrous outer membrane and nerve invasion were found to be significantly associated with poorer overall survival, consistent with previous studies that have identified these pathological features as independent adverse prognostic indicators(19, 23). This association may be attributed to several underlying mechanisms. The presence of nerve invasion and fibrous outer membrane invasion reflects deeper infiltration of tumor cells into surrounding anatomical structures. These features typically occur in more advanced disease stages and signify enhanced tumor aggressiveness and local invasiveness. Nerve invasion is often associated with perineural spread, allowing tumor cells to disseminate along nerve sheaths (24). Fibrous outer membrane invasion suggests that the tumor has breached the anatomical confines of the esophageal wall, increasing the likelihood of local recurrence or distant dissemination. Since this study focused on patients already exhibiting vascular invasion—a known high-risk feature—the coexistence of NI and outer membrane invasion further compounds the metastatic potential. This multifocal invasive behavior may indicate a biologically aggressive subtype prone to early recurrence and treatment resistance. Tumors infiltrating perineural or fibrous structures often reside in complex tissue interfaces, which may promote immune evasion and foster a more permissive microenvironment for progression(25). Such settings can impair treatment efficacy and contribute to poorer outcomes. Therefore, recognizing these features can support risk stratification and guide the implementation of more aggressive surveillance and therapeutic strategies. Additionally, both T and N stages were independently associated with overall survival(26). Specifically, advanced T stage reflects deeper tumor infiltration through the esophageal wall, which is often linked to greater tumor burden and increased likelihood of adjacent structure involvement (27). Similarly, higher N stage indicates more extensive regional lymph node metastasis, suggesting aggressive tumor biology and a higher risk of systemic dissemination(28). Taken together, these findings are also consistent with established oncologic principles and highlight the prognostic significance of tumor depth and nodal involvement even within this high-risk subgroup. Notably, in our study, only T stage ≥ T3 showed a significant prognostic impact, underscoring that T2-stage tumors with vascular invasion may not independently confer limited OS, and suggesting that, while vascular invasion is an adverse feature, its prognostic effect may be modulated by the depth of tumor infiltration. In our study, lower BMI was significantly associated with poorer overall survival. This inverse relationship suggests that undernutrition or cancer-related cachexia may contribute to worse outcomes in this high-risk population. Malnutrition is known to impair immune function, delay postoperative recovery, and reduce tolerance to chemotherapy or radiotherapy(29, 30). Furthermore, a low BMI may reflect advanced disease burden and systemic inflammation, both of which are linked to unfavorable prognosis(31). These factors collectively compromise the patient's physiological reserve and treatment responsiveness, ultimately contributing to poorer overall survival. Additionally, elevated WBC count was also associated with poorer overall survival in our study, suggesting that systemic inflammation may significantly contribute to tumor progression and adverse outcomes. Increased WBC levels reflect a heightened inflammatory state, which can promote tumor growth, angiogenesis, and metastasis through the release of cytokines and other mediators (32). In parallel, chronic inflammation may also induce oxidative stress and DNA damage, further accelerating tumor evolution. Moreover, inflammation-driven immune dysregulation may impair the body's ability to mount an effective anti-tumor response, weakening host defense and facilitating disease progression (33). Thus, our findings and the above mechanisms indicate that elevated WBC is both a marker of systemic inflammation and a predictor of poor prognosis in this high-risk subgroup. Our findings also indicated that SCC level was one of the key factors affecting overall survival in patients with vascular invasion. Mechanistically, SCC, a serological tumor marker secreted by squamous epithelial cells, has been associated with enhanced tumor cell proliferation, inhibition of apoptosis, epithelial-mesenchymal transition (EMT), and immune evasion(34). These biological processes contribute to increased tumor invasiveness and metastatic potential. Furthermore, in the context of vascular invasion, elevated SCC levels may indicate a more aggressive tumor phenotype characterized by increased angiogenesis, vascular permeability, and the potential for hematogenous dissemination. High SCC levels could reflect the extent of vascular involvement and the presence of micrometastases not captured by conventional imaging or staging systems. Therefore, SCC not only reflects disease severity but may also actively participate in tumor progression, justifying its prognostic value in patients with vascular invasion. A notable finding in this analysis was that a higher total number of lymph nodes dissected was significantly associated with poorer overall survival in esophageal cancer patients with vascular invasion. While this finding may appear counterintuitive, it likely reflects the biological aggressiveness and extensive regional spread of tumors in this subgroup. A greater number of dissected lymph nodes often corresponds to more advanced disease with widespread microscopic metastases or pronounced lymphovascular infiltration, both of which portend a poor prognosis. Moreover, excessive lymphadenectomy may disrupt the local lymphatic architecture and immune surveillance, fostering a tumor-permissive microenvironment characterized by immune suppression and chronic inflammation(35). Such changes can impair the body’s antitumor immunity and promote residual tumor cell survival and dissemination. Additionally, extensive lymph node dissection may lead to increased postoperative complications and systemic inflammatory responses, further exacerbating patient outcomes(36). These findings underscore the importance of precise nodal staging and judicious lymphadenectomy. While adequate lymph node evaluation is essential for accurate prognosis and treatment planning, overtly aggressive dissection in biologically advanced disease may confer limited benefit and potential harm. Tailored surgical strategies that balance oncologic clearance with preservation of immune integrity are warranted in this high-risk population. Overall, our study successfully established a robust and interpretable prognostic model using machine learning techniques, with the GBM model demonstrating strong discrimination, calibration, and clinical utility in esophageal cancer patients with vascular invasion. Despite the promising results, this study has several limitations that warrant consideration. First, the study was conducted retrospectively and based on data from a single institution, which may introduce selection bias and limit the generalizability of the findings. Although rigorous internal validation was performed, the robustness and applicability of the GBM model should be further confirmed through prospective validation using larger, independent, and multicenter cohorts that encompass broader demographic and clinical heterogeneity. Second, while machine learning algorithms such as GBM demonstrated superior predictive performance, one of the enduring criticisms of such models lies in their limited interpretability. Although GBM offers a better balance between accuracy and interpretability compared to more opaque models like deep neural networks, the clinical adoption of such tools depends on transparency and trust. Future work incorporating model explanation techniques, such as SHapley Additive exPlanations (SHAP), would be valuable in elucidating the contribution of individual features to survival predictions, thus enhancing interpretability and clinician confidence. Third, our model focused exclusively on OS as the primary outcome. While OS remains a fundamental endpoint in oncologic prognostication, additional endpoints such as DFS, PFS, and response to neoadjuvant or adjuvant therapies are equally critical in optimizing treatment strategies for esophageal cancer. Future studies should aim to develop and validate models tailored to these outcomes, potentially enabling more nuanced and comprehensive clinical decision-making. Lastly, this study did not incorporate certain potentially important variables—such as molecular biomarkers, detailed nutritional assessments, and specific treatment parameters including radiotherapy dosage and chemotherapy protocols—owing to limitations in data availability. Future studies that include these additional prognostic factors may contribute to the development of more comprehensive and accurate predictive models with broader clinical applicability. Conclusion This study developed and validated a machine learning–based prognostic model for esophageal cancer patients with vascular invasion, a subgroup known for poor outcomes. By integrating multiple clinical, pathological, and laboratory features, the GBM model demonstrated superior predictive performance in terms of discrimination, calibration, and clinical utility. The model effectively stratified patients into distinct risk groups, providing a reliable tool for individualized survival prediction. Importantly, several variables—including nerve invasion, invasion of the fibrous outer membrane, T and N stage, BMI, WBC, SCC levels, and total number of dissected lymph nodes—were identified as key predictors and offer mechanistic insights into disease progression. These findings underscore the potential of machine learning approaches to enhance risk assessment and guide treatment decision-making in esophageal cancer patients with vascular invasion. Future multicenter prospective studies and external validations are warranted to further confirm the generalizability and clinical applicability of this model. Abbreviations EC: Esophageal Cancer VI: Vascular Invasion PFS: Progression-Free Survival DFS: Disease-Free Survival OS: Overall Survival ML: Machine Learning BMI: Body Mass Index AJCC: American Joint Committee on Cancer CT: Computed Tomography HE: Hematoxylin and Eosin WHO: World Health Organization TC: Training Cohort VC: Validation Cohort SD: Standard Deviation RSF: Random Survival Forest GBM: Gradient Boosting Machine LASSO-Cox: Least Absolute Shrinkage and Selection Operator Cox Regression CoxBoost: Cox Proportional Hazards Model with Boosting Survivalsvm: Survival Support Vector Machine XGBoost: Extreme Gradient Boosting SuperPC: Supervised Principal Components PlsRcox: Partial Least Squares Regression for Cox Time-AUC: Time-dependent Area Under the Curve AUC: Area Under the Curve ROC: Receiver Operating Characteristic DCA: Decision Curve Analysis WBC: White Blood Cell count PLT: Platelet count SA: Serum Sialic Acid SCC: Squamous Cell Carcinoma antigen SF: Serum Ferritin NI: Nerve Invasion HR: Hazard Ratio CI: Confidence Interval EMT: Epithelial-Mesenchymal Transition SHAP: SHapley Additive exPlanations Declarations Ethics approval and consent to participate This retrospective study was approved by the Ethics Committee of Qilu Hospital, Shandong University (Approval number: KYLL-202008-023-1) and conducted in accordance with the ethical standards of the Declaration of Helsinki. Given the retrospective design and use of anonymized clinical data, the requirement for written informed consent was waived by the ethics committee. Consent for Publication This manuscript does not contain any identifiable personal data of individual participants in any form. Availability of Data and Materials Access to the datasets analyzed in this study is available from the corresponding author on reasonable request. Competing Interests There are no conflicts of interest to disclose. Funding This work was supported by the Taishan Scholar Program of Shandong Province (ts201712087) and the National Natural Science Foundation of China (Grant No. 82472814). Author contributions YZ and JJL conceived and designed the study, performed data collection and statistical analysis, developed the machine learning models, and drafted the manuscript. ZZ, ZYL, YJM, SJZ, and HML contributed to data acquisition, preprocessing, and result interpretation. JHQ assisted in model validation and manuscript formatting. HT supervised the project, provided methodological and conceptual guidance, and critically revised the manuscript. All authors read and approved the final version of the manuscript. Acknowledgements We extend our sincere gratitude to the patients and medical staff of Qilu Hospital of Shandong University for their invaluable contributions to this study. Special thanks are also due to the radiologists and pathologists for their expert support in data interpretation and analysis. References Li N, Chen S, Wang X, Zhang B, Zeng B, Sun C, et al. Identification of POU4F1 as a novel prognostic biomarker and therapeutic target in esophageal squamous cell carcinoma. Cancer Cell Int. 2024;24(1):280. Sheikh M, Roshandel G, McCormack V, Malekzadeh R. Current Status and Future Prospects for Esophageal Cancer. Cancers (Basel). 2023;15(3). Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, 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):209–49. Mulisa G, Pero-Gascon R, McCormack V, Bisanz JE, Talukdar FR, Abebe T, et al. Multiple mycotoxin exposure assessment through human biomonitoring in an esophageal cancer case-control study in the Arsi-Bale districts of Oromia region of Ethiopia. Int J Hyg Environ Health. 2025;263:114466. Gonzalez J, Bahmad HF, Ocejo S, Abreu A, Popp M, Gogola S, et al. The Usefulness of Elastin Staining to Detect Vascular Invasion in Cancer. Int J Mol Sci. 2023;24(20). Ozer L, Senocak Tasci E, Mutlu AU, Piyade B, Ramoglu N, Ajredini M, et al. Intramural Component of Venous, Lymphatic, and Perineural Invasion in Colon Cancer: A Threat or an Illusion? Balkan Med J. 2022;39(6):436–43. Lin Y, Zhang Y, Fang H, Hu Q, Duan H, Zhang L, et al. Survival and clinicopathological significance of blood vessel invasion in operable breast cancer: a systematic review and meta-analysis. Jpn J Clin Oncol. 2023;53(1):35–45. Engholm G, Lundberg FE, Konig SM, Olafsdottir E, Johannesen TB, Pettersson D, et al. TNM stage in the Nordic Cancer Registries 2004-2016: Registration and availability. Acta Oncol. 2024;63:303–12. Wang A, Tan Y, Wang S, Chen X. The prognostic value of separate lymphatic invasion and vascular invasion in oesophageal squamous cell carcinoma: a meta-analysis and systematic review. BMC Cancer. 2022;22(1):1329. Corachea AJM, Ferrer RJE, Ty LPB, Aquino LAC, Morta MT, Macalindong SS, et al. Lymphovascular Invasion Is Associated With Doxorubicin Resistance in Breast Cancer. Lab Invest. 2025;105(5):104115. Xie C, Chen Z, Xu J, Meng Z, Huang Z, Lin J. Influence of Lymphangio vascular (V) and perineural (N) invasion on survival of patients with resected esophageal squamous cell carcinoma (ESCC): a single-center retrospective study. PeerJ. 2022;10:e12974. Oguma J, Ishiyama K, Kurita D, Kanematsu K, Kubo K, Utsunomiya D, et al. Oncologic significance of lymphovascular invasion in patients with superficial esophageal squamous cell carcinoma reaching the muscularis mucosae or with slight invasion of the submucosa. Surgery. 2023;173(5):1162–8. Sakkal M, Hajal AA. Machine learning predictions of tumor progression: How reliable are we? Comput Biol Med. 2025;191:110156. Fiste O, Liontos M, Zagouri F, Stamatakos G, Dimopoulos MA. Machine learning applications in gynecological cancer: A critical review. Crit Rev Oncol Hematol. 2022;179:103808. Hao Z, Wang Z, Ma J, Li Y, Zhang W. Harnessing the machine learning and nomogram models: elevating prognostication in nonmetastatic gastric cancer with "double invasion" for personalized patient care. Eur J Med Res. 2025;30(1):517. Mohammadian Rad N, Sosef O, Seegers J, Koolen L, Hoofwijk J, Woodruff HC, et al. Prognostic models for colorectal cancer recurrence using carcinoembryonic antigen measurements. Front Oncol. 2024;14:1368120. Huang X, Wang Q, Xu W, Liu F, Pan L, Jiao H, et al. Machine learning to predict lymph node metastasis in T1 esophageal squamous cell carcinoma: a multicenter study. Int J Surg. 2024;110(12):7852–9. Teja R, Bhaskar R, Yadlapalli DC, Kukunuri S, Mohan R. Real-World Data of Esophageal Malignancies Managed Under the Government Scheme From a Tertiary Cancer Center in South India. Cureus. 2025;17(1):e77850. Zhou J, Yang Y, Zhang H, Luan S, Xiao X, Li X, et al. Lymphovascular and Perineural Invasion After Neoadjuvant Therapy in Esophageal Squamous Carcinoma. Ann Thorac Surg. 2023;115(6):1386–94. Oguma J, Ishiyama K, Kurita D, Kanematsu K, Kubo K, Utsunomiya D, et al. Significance of lymphovascular invasion in esophageal squamous cell carcinoma undergoing neoadjuvant chemotherapy followed by esophagectomy. Esophagus. 2023;20(2):215–24. Salditt M, Humberg S, Nestler S. Gradient Tree Boosting for Hierarchical Data. Multivariate Behav Res. 2023;58(5):911–37. Zhang X, Xue Y, Su X, Chen S, Liu K, Chen W, et al. A Transfer Learning Approach to Correct the Temporal Performance Drift of Clinical Prediction Models: Retrospective Cohort Study. JMIR Med Inform. 2022;10(11):e38053. Matsumoto T, Noma K, Maeda N, Kato T, Moriwake K, Kawasaki K, et al. Safe and curative modified two-stage operation for T4 esophageal cancer after definitive chemoradiotherapy: a case report. Surg Case Rep. 2023;9(1):119. Zhu M, Luo F, Xu B, Xu J. Research Progress of Neural Invasion in Pancreatic Cancer. Curr Cancer Drug Targets. 2024;24(4):397–410. Ma C, Luo H. A more novel and robust gene signature predicts outcome in patients with esophageal squamous cell carcinoma. Clin Res Hepatol Gastroenterol. 2022;46(10):102033. Krauss DT, Schmidt T, Bruns CJ, Fuchs HF. [Evidence for the extent and oncological benefit of lymphadenectomy for esophageal cancer]. Chirurgie (Heidelb). 2025;96(4):273–80. Yu Y, Wei X, Chen X, Li H, Liu Q, Sun H, et al. The T stage of esophageal cancer can be effectively predicted by muscularis propria thickness and muscularis propria + mucosa thickness under ultrasonic gastroscopy. Thorac Cancer. 2023;14(2):127–34. Wang Z, Li F, Zhu M, Lu T, Wen L, Yang S, et al. Prognostic prediction and comparison of three staging programs for patients with advanced (T2-T4) esophageal squamous carcinoma after radical resection. Front Oncol. 2024;14:1376527. Nakayama K, Yoshida T, Nakayama Y, Iguchi N, Namba Y, Konishi M, et al. Activation of macrophages mediates dietary restriction-induced splenic involution. Life Sci. 2022;310:121068. Santos SS, Costa L, Araripe TSO, Reges B, Ximenes HMA, Moreira A. Immunomodulatory enteral nutrition in post-surgical gastrointestinal cancer: Clinical, biochemical and nutritional impacts. Clin Nutr ESPEN. 2025;68:254–62. Locks LM, Parekh A, Newell K, Dauphinais MR, Cintron C, Maloomian K, et al. The ABCDs of Nutritional Assessment in Infectious Diseases Research. J Infect Dis. 2025;231(3):562–72. Fruntelata RF, Bakri A, Stoica GA, Mogoanta L, Ionovici N, Popescu G, et al. Assessment of tumoral and peritumoral inflammatory reaction in cutaneous malignant melanomas. Rom J Morphol Embryol. 2023;64(1):41–8. Jalali AM, Mitchell KJ, Pompoco C, Poludasu S, Tran S, Ramana KV. Therapeutic Significance of NLRP3 Inflammasome in Cancer: Friend or Foe? Int J Mol Sci. 2024;25(24). Hsu LC, Lin CN, Hsu FT, Chen YT, Chang PL, Hsieh LL, et al. Imipramine Suppresses Tumor Growth and Induces Apoptosis in Oral Squamous Cell Carcinoma: Targeting Multiple Processes and Signaling Pathways. Anticancer Res. 2023;43(9):3987–96. Patel AJ, Bille A. Lymph node dissection in lung cancer surgery. Front Surg. 2024;11:1389943. Baud G, Jannin A, Marciniak C, Chevalier B, Do Cao C, Leteurtre E, et al. Impact of Lymph Node Dissection on Postoperative Complications of Total Thyroidectomy in Patients with Thyroid Carcinoma. Cancers (Basel). 2022;14(21). Tables Table 1. Baseline characteristics of patients stratified by survival status in the training and validation cohorts. TC VC TC vs VC Alive (n=110) Dead (n=113) p value Alive (n=50) Dead (n=45) p value p value Age (mean (SD)) 63.27 (8.13) 63.17 (7.89) 0.922 63.54 (9.40) 64.96 (8.14) 0.437 0.327 Gender (%) 0.41 1 0.774 Male 90 (81.8) 98 (86.7) 43 (86.0) 39 (86.7) Female 20 (18.2) 15 (13.3) 7 (14.0) 6 (13.3) Weight (mean (SD)) 66.42 (10.37) 64.73 (10.85) 0.235 64.18 (10.53) 61.82 (9.73) 0.261 0.053 BMI (mean (SD)) 24.33 (2.91) 23.51 (3.34) 0.051 23.47 (3.28) 23.17 (2.86) 0.63 0.126 Smoking history (%) 1 0.16 0.263 No 40 (36.4) 42 (37.2) 26 (52.0) 16 (35.6) Yes 70 (63.6) 71 (62.8) 24 (48.0) 29 (64.4) History of alcohol intake (%) 0.569 0.665 0.694 No 42 (38.2) 38 (33.6) 21 (42.0) 16 (35.6) Yes 68 (61.8) 75 (66.4) 29 (58.0) 29 (64.4) Hypertension (%) 0.168 0.215 0.958 No 85 (77.3) 77 (68.1) 40 (80.0) 30 (66.7) Yes 25 (22.7) 36 (31.9) 10 (20.0) 15 (33.3) Diabetes (%) 0.936 0.279 0.984 No 94 (85.5) 98 (86.7) 45 (90.0) 36 (80.0) Yes 16 (14.5) 15 (13.3) 5 (10.0) 9 (20.0) Tumor size (mean (SD)) 4.76 (1.85) 5.51 (2.18) 0.006* 4.68 (1.75) 4.98 (2.02) 0.478 0.189 Tumor location (%) 0.493 0.208 Upper esophagus 5 (4.5) 1 (0.9) 3 (6.0) 1 (2.2) Middle esophagus 20 (18.2) 23 (20.4) 8 (16.0) 11 (24.4) Lower esophagus 53 (48.2) 53 (46.9) 28 (56.0) 19 (42.2) Upper and middle esophagus 3 (2.7) 5 (4.4) 3 (6.0) 5 (11.1) Middle and lower esophagus 29 (26.4) 31 (27.4) 8 (16.0) 9 (20.0) 0.444 Invasion of fibrous outer membrane (%) <0.001* <0.001* 0.812 No 85 (77.3) 51 (45.1) 41 (82.0) 19 (42.2) Yes 25 (22.7) 62 (54.9) 9 (18.0) 26 (57.8) Hemoglobin (mean (SD)) 136.40 (16.16) 136.51 (18.20) 0.961 139.90 (17.23) 140.04 (15.79) 0.966 0.092 WBC (mean (SD)) 5.73 (1.71) 6.46 (1.60) 0.001* 5.77 (1.58) 6.15 (1.77) 0.265 0.466 NeutroCount (mean (SD)) 3.42 (1.41) 4.55 (6.00) 0.055 3.53 (1.46) 3.82 (1.66) 0.378 0.485 LymphoCount (mean (SD)) 1.71 (0.61) 1.72 (0.55) 0.855 1.66 (0.47) 1.66 (0.43) 0.964 0.404 MonoCount (mean (SD)) 0.48 (0.22) 0.54 (0.20) 0.033 0.43 (0.13) 0.51 (0.18) 0.013* 0.137 PLT (mean (SD)) 232.12 (66.24) 248.69 (72.87) 0.077 224.34 (58.86) 252.71 (73.10) 0.039* 0.747 Total protein (mean (SD)) 68.75 (5.09) 68.44 (4.72) 0.634 68.06 (5.13) 68.92 (9.15) 0.57 0.861 Albumin (mean (SD)) 43.17 (3.27) 43.34 (3.43) 0.703 42.79 (3.39) 44.13 (2.80) 0.041* 0.681 Prealbumin (mean (SD)) 22.95 (4.63) 22.88 (4.80) 0.917 23.54 (3.94) 23.68 (5.16) 0.881 0.223 Tbil (mean (SD)) 11.52 (4.82) 11.17 (4.75) 0.587 12.12 (3.76) 11.50 (3.72) 0.417 0.379 Dbil (mean (SD)) 6.89 (31.41) 3.89 (1.50) 0.312 3.94 (1.17) 4.07 (1.32) 0.616 0.547 ALT (mean (SD)) 14.82 (9.92) 16.15 (12.16) 0.372 15.90 (6.65) 17.31 (16.83) 0.585 0.447 AST (mean (SD)) 18.56 (6.18) 18.96 (8.38) 0.692 20.00 (6.84) 19.67 (9.48) 0.844 0.247 BUN (mean (SD)) 5.40 (1.99) 5.39 (1.54) 0.946 5.23 (1.32) 5.48 (1.55) 0.397 0.807 Scr (mean (SD)) 75.54 (29.41) 75.30 (15.35) 0.938 74.58 (13.54) 76.11 (16.12) 0.616 0.964 Glucose (mean (SD)) 6.17 (8.54) 5.30 (1.09) 0.288 5.55 (1.21) 5.69 (2.11) 0.691 0.862 Cholesterol (mean (SD)) 4.75 (0.86) 4.75 (1.06) 0.979 4.77 (0.97) 4.78 (0.96) 0.978 0.842 HDL (mean (SD)) 1.25 (0.27) 1.25 (0.29) 0.938 1.25 (0.35) 1.27 (0.26) 0.777 0.713 LDL (mean (SD)) 2.86 (0.70) 2.92 (0.79) 0.515 2.89 (0.65) 2.85 (0.77) 0.816 0.851 Triglyceride (mean (SD)) 1.18 (0.64) 1.16 (0.51) 0.783 1.34 (0.71) 1.32 (0.66) 0.862 0.032* Uric acid (mean (SD)) 299.44 (60.50) 302.15 (69.57) 0.756 308.76 (80.38) 303.13 (79.27) 0.732 0.537 LDH (mean (SD)) 193.28 (36.08) 197.84 (33.25) 0.327 192.22 (34.29) 196.07 (33.82) 0.584 0.714 SA (mean (SD)) 59.17 (9.19) 61.70 (9.04) 0.039* 55.94 (8.08) 60.09 (10.05) 0.028* 0.024* CEA (mean (SD)) 3.15 (3.90) 3.29 (2.34) 0.75 2.59 (1.73) 2.98 (1.60) 0.269 0.196 AFP (mean (SD)) 3.22 (1.55) 2.84 (1.24) 0.046* 3.07 (1.65) 2.97 (1.21) 0.734 0.961 SCC (mean (SD)) 0.97 (0.64) 1.61 (1.99) 0.002* 1.15 (1.05) 1.55 (1.30) 0.105 0.785 SF (mean (SD)) 218.18 (194.72) 279.40 (169.45) 0.013* 227.93 (177.56) 259.82 (160.24) 0.363 0.78 CA199 (mean (SD)) 16.73 (33.37) 12.00 (13.96) 0.167 12.70 (8.05) 12.20 (8.49) 0.773 0.485 CA125 (mean (SD)) 9.98 (5.51) 9.32 (4.22) 0.313 9.58 (4.16) 10.25 (4.40) 0.452 0.664 CA724 (mean (SD)) 4.60 (9.81) 3.51 (3.15) 0.264 3.92 (4.27) 2.71 (1.33) 0.071 0.371 PNI (mean (SD)) 51.70 (4.95) 52.23 (5.55) 0.455 51.65 (5.66) 52.41 (3.98) 0.456 0.955 SII (mean (SD)) 565.55 (616.91) 648.78 (408.90) 0.235 543.85 (382.70) 626.49 (411.74) 0.313 0.68 Surgical approach (%) 0.367 0.469 0.947 McKeown 49 (44.5) 50 (44.2) 22 (44.0) 21 (46.7) Ivor Lewis 2 (1.8) 6 (5.3) 1 (2.0) 3 (6.7) Sweet 59 (53.6) 57 (50.4) 27 (54.0) 21 (46.7) Surgical method (%) 0.788 0.314 0.59 Minimally invasive surgery 46 (41.8) 44 (38.9) 21 (42.0) 19 (42.2) Robotic surgery 4 (3.6) 6 (5.3) 0 (0.0) 2 (4.4) Open surgery 60 (54.5) 63 (55.8) 29 (58.0) 24 (53.3) Neoadjuvant therapy (%) 0.938 0.824 0.631 No 95 (86.4) 99 (87.6) 43 (86.0) 37 (82.2) Yes 15 (13.6) 14 (12.4) 7 (14.0) 8 (17.8) Pathological type (%) 0.326 0.694 0.562 ESCC 88 (80.0) 97 (85.8) 42 (84.0) 40 (88.9) EAC 22 (20.0) 16 (14.2) 8 (16.0) 5 (11.1) Differentiation grade (%) 0.117 0.053 0.92 WD 7 (6.4) 5 (4.4) 6 (12.0) 0 (0.0) MD 53 (48.2) 41 (36.3) 21 (42.0) 20 (44.4) LD 50 (45.5) 67 (59.3) 23 (46.0) 25 (55.6) T stage (%) <0.001* 0.023* 0.455 T1 19 (17.3) 2 (1.8) 8 (16.0) 3 (6.7) T2 36 (32.7) 16 (14.2) 19 (38.0) 9 (20.0) T3 53 (48.2) 84 (74.3) 23 (46.0) 30 (66.7) T4 2 (1.8) 11 (9.7) 0 (0.0) 3 (6.7) N stage (%) <0.001* 0.008* 0.327 N0 46 (41.8) 20 (17.7) 16 (32.0) 10 (22.2) N1 37 (33.6) 37 (32.7) 24 (48.0) 11 (24.4) N2 19 (17.3) 33 (29.2) 8 (16.0) 19 (42.2) N3 8 (7.3) 23 (20.4) 2 (4.0) 5 (11.1) NI (%) <0.001* <0.001* 0.309 No 91 (82.7) 60 (53.1) 41 (82.0) 17 (37.8) Yes 19 (17.3) 53 (46.9) 9 (18.0) 28 (62.2) Total number of lymph nodes (mean (SD)) 17.64 (7.93) 20.26 (8.84) 0.021* 17.30 (7.97) 16.44 (8.31) 0.61 0.045* Abbreviations: TC, Training Cohort; VC, Validation Cohort; BMI, Body Mass Index; WBC, White Blood Cell count; PLT, Platelet count; TBil, Total Bilirubin; DBil, Direct Bilirubin; ALT, Alanine Aminotransferase; AST, Aspartate Aminotransferase; BUN, Blood Urea Nitrogen; Scr, Serum Creatinine; HDL, High-Density Lipoprotein; LDL, Low-Density Lipoprotein; LDH, Lactate Dehydrogenase; SA, Serum Sialic Acid; CEA, Carcinoembryonic Antigen; AFP, Alpha-Fetoprotein; SCC, Squamous Cell Carcinoma antigen; SF, Serum Ferritin; CA199, Carbohydrate Antigen 199; CA125, Carbohydrate Antigen 125; CA724, Carbohydrate Antigen 724; PNI, Prognostic Nutritional Index; SII, Systemic Immune-Inflammation Index; ESCC, Esophageal Squamous Cell Carcinoma; EAC, Esophageal Adenocarcinoma; WD, Well Differentiated; MD, Moderately Differentiated; LD, Low Differentiation; NI, Nerve Invasion; SD, Standard Deviation. Table2 Univariate and multivariate Cox regression analyses identifying prognostic factors associated with overall survival. Univariate analysis Multivariate analysis OR (95%CI) P value OR (95%CI) P value Age 1.00 (0.98-1.02) 0.865 Gender Male Ref Female 0.77 (0.44-1.32) 0.337 Weight 0.99 (0.97-1.00) 0.129 BMI 0.93 (0.87-0.99) 0.016* 0.93 (0.87-0.99) 0.019* Smoking history No Ref Yes 0.95 (0.65-1.39) 0.782 History of alcohol intake No Ref Yes 1.05 (0.71-1.56) 0.791 Hypertension No Ref Yes 1.35 (0.91-2.00) 0.138 Diabetes No Ref Yes 0.94 (0.54-1.61) 0.809 Tumor size 1.14 (1.05-1.24) 0.002* 1.02 (0.92-1.14) 0.682 Tumor location Upper esophagus Ref Middle esophagus 3.64 (0.49-26.94) 0.206 Lower esophagus 3.56 (0.49-25.78) 0.208 Upper and middle esophagus 5.10 (0.60-43.71) 0.137 Middle and lower esophagus 3.80 (0.52-27.86) 0.189 Invasion of fibrous outer membrane No Ref Yes 2.66 (1.83-3.86) <0.001* 1.74 (1.14-2.65) 0.011* Hemoglobin 1.00 (0.99-1.01) 0.917 WBC 1.18 (1.07-1.30) 0.001* 1.19 (1.06-1.34) 0.002* NeutroCount 1.02 (1.00-1.05) 0.089 LymphoCount 1.01 (0.74-1.38) 0.994 MonoCount 1.91 (0.99-3.66) 0.053 PLT 1.00 (1.00-1.01) 0.061 Total protein 0.99 (0.95-1.02) 0.505 Albumin 1.00 (0.95-1.06) 0.960 Prealbumin 0.99 (0.95-1.03) 0.638 Tbil 0.99 (0.95-1.03) 0.589 Dbil 0.99 (0.92-1.05) 0.672 ALT 1.00 (0.99-1.02) 0.692 AST 1.00 (0.97-1.03) 0.962 BUN 1.00 (0.91-1.11) 0.974 Scr 1.00 (0.99-1.01) 0.992 Glucose 0.97 (0.88-1.06) 0.491 Cholesterol 0.99 (0.81-1.21) 0.902 HDL 1.04 (0.54-2.02) 0.903 LDL 1.06 (0.82-1.38) 0.653 Triglyceride 0.93 (0.66-1.31) 0.682 Uric acid 1.00 (1.00-1.00) 0.920 LDH 1.00 (1.00-1.01) 0.246 SA 1.02 (1.00-1.04) 0.028* 1.00 (0.98-1.02) 0.856 CEA 1.00 (0.96-1.06) 0.848 AFP 0.87 (0.75-1.02) 0.081 SCC 1.11 (1.04-1.18) <0.001* 1.12 (1.04-1.20) 0.002* SF 1.00 (1.00-1.00) 0.025* 1.00 (1.00-1.00) 0.264 CA199 0.99 (0.98-1.00) 0.208 CA125 0.98 (0.95-1.02) 0.440 CA724 0.98 (0.95-1.02) 0.368 PNI 1.01 (0.97-1.05) 0.582 SII 1.00 (1.00-1.00) 0.202 Surgical approach McKeown Ref Ivor Lewis 1.73 (0.74-4.04) 0.204 Sweet 0.97 (0.66-1.41) 0.863 Surgical method Minimally invasive surgery Ref Robotic surgery 1.42 (0.61-3.34) 0.416 Open surgery 1.07 (0.73-1.58) 0.717 Neoadjuvant therapy No Ref Yes 0.87 (0.49-1.52) 0.615 Pathological type ESCC Ref EAC 0.76 (0.45-1.28) 0.300 Differentiation grade WD Ref MD 1.10 (0.44-2.79) 0.836 LD 1.62 (0.65-4.03) 0.296 T stage T1 Ref T2 3.94 (0.91-17.15) 0.067 2.80 (0.63-12.42) 0.175 T3 9.40 (2.31-38.24) 0.002* 4.65 (1.09-19.87) 0.038* T4 24.11 (5.33-109.09) <0.001* 6.31 (1.27-31.28) 0.024* N stage N0 Ref N1 1.99 (1.16-3.43) 0.013* 1.77 (1.00-3.11) 0.049* N2 3.01 (1.72-5.25) <0.001* 1.98 (1.08-3.61) 0.027* N3 4.27 (2.34-7.80) <0.001* 2.43 (1.25-4.73) 0.009* NI No Ref Yes 2.49 (1.72-3.61) <0.001* 1.54 (1.01-2.34) 0.043* Total number of lymph nodes 1.03 (1.01-1.06) 0.002* 1.03 (1.00-1.05) 0.022* Abbreviations: OR, Odds Ratio; Ref, Reference; BMI, Body Mass Index; WBC, White Blood Cell count; PLT, Platelet count; TBil, Total Bilirubin; DBil, Direct Bilirubin; ALT, Alanine Aminotransferase; AST, Aspartate Aminotransferase; BUN, Blood Urea Nitrogen; Scr, Serum Creatinine; HDL, High-Density Lipoprotein; LDL, Low-Density Lipoprotein; LDH, Lactate Dehydrogenase; SA, Serum Sialic Acid; CEA, Carcinoembryonic Antigen; AFP, Alpha-Fetoprotein; SCC, Squamous Cell Carcinoma antigen; SF, Serum Ferritin; CA199, Carbohydrate Antigen 199; CA125, Carbohydrate Antigen 125; CA724, Carbohydrate Antigen 724; PNI, Prognostic Nutritional Index; SII, Systemic Immune-Inflammation Index; ESCC, Esophageal Squamous Cell Carcinoma; EAC, Esophageal Adenocarcinoma; WD, Well Differentiated; MD, Moderately Differentiated; LD, Low Differentiation; NI, Nerve Invasion; Additional Declarations No competing interests reported. Supplementary Files Additionalfile1.tif Additional file 1 Calibration curves at 1-, 2-, and 3-year time points for seven machine learning models: RSF, LASSO-Cox, CoxBoost, SurvivalSVM, XGBoost, SuperPC, and plsRcox. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 01 Dec, 2025 Reviewers agreed at journal 26 Nov, 2025 Reviewers invited by journal 12 Nov, 2025 Editor invited by journal 23 Jul, 2025 Editor assigned by journal 21 Jul, 2025 Submission checks completed at journal 18 Jul, 2025 First submitted to journal 17 Jul, 2025 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-7151065","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":549146881,"identity":"1fe686f0-e3fa-4468-8f0b-68912f791dd4","order_by":0,"name":"Yue Zhao","email":"","orcid":"","institution":"Qilu Hospital of Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Yue","middleName":"","lastName":"Zhao","suffix":""},{"id":549146882,"identity":"d830761f-95fb-4968-8b41-bcd96f96d76b","order_by":1,"name":"Jun-jie Liu","email":"","orcid":"","institution":"Qilu Hospital of Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Jun-jie","middleName":"","lastName":"Liu","suffix":""},{"id":549146883,"identity":"2152e6a6-36a0-4b56-afa1-26858b7117a9","order_by":2,"name":"zhan zhang","email":"","orcid":"","institution":"Qianfoshan Hospital in the Shandong Province","correspondingAuthor":false,"prefix":"","firstName":"zhan","middleName":"","lastName":"zhang","suffix":""},{"id":549146884,"identity":"ecb0e3ce-3673-4420-847b-79805393e02c","order_by":3,"name":"Zhen-yi Li","email":"","orcid":"","institution":"Qilu Hospital of Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Zhen-yi","middleName":"","lastName":"Li","suffix":""},{"id":549146885,"identity":"a21d3ca5-9181-4431-9a46-dacfc6985bb8","order_by":4,"name":"Yi-jun Ma","email":"","orcid":"","institution":"Qilu Hospital of Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Yi-jun","middleName":"","lastName":"Ma","suffix":""},{"id":549146886,"identity":"be13916f-7c7e-42fd-906f-34942f66cd9d","order_by":5,"name":"Si-jie Zhang","email":"","orcid":"","institution":"Qilu Hospital of Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Si-jie","middleName":"","lastName":"Zhang","suffix":""},{"id":549146887,"identity":"949ac827-13ac-4401-9ccf-52d200c4acfa","order_by":6,"name":"Hai-ming Li","email":"","orcid":"","institution":"Qilu Hospital of Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Hai-ming","middleName":"","lastName":"Li","suffix":""},{"id":549146888,"identity":"1eac76ff-4f2a-4a41-a3ef-b75a659fb1a9","order_by":7,"name":"Jian-hao Qiu","email":"","orcid":"","institution":"Qianfoshan Hospital in the Shandong Province","correspondingAuthor":false,"prefix":"","firstName":"Jian-hao","middleName":"","lastName":"Qiu","suffix":""},{"id":549146889,"identity":"aec1f4cc-fce4-44d4-a7d9-6c8016a08a0f","order_by":8,"name":"Hui Tian","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABDElEQVRIiWNgGAWjYDACZjCSYOBjbwDyDCyARAIQs+HWwQPTwsZzAKRFgggtEIuAaiRAKhmI0GLPzvzscWGbhWyb5OOjG34USDDIt+cYMHwoO8zAP7sBh8PYzI1ntkkYt0mnpd3sATrM4MwbA8YZ5w4zSNw5gMsvZtK8bRKJbdI5Zjd4QFokcgyYedsOAxkJOLSwf4NokTz/7eYfoBb5GUAtf/Fq4YHaIsHDdhtkC8MNoBZGfFoO85RJ85wD+oUnzey2jIEEj8GZZwUHe86l80jcwK6Fvf/4NmmesjrZfvbDz26++WMjJ9+evPHBjzJrOf4Z2LXAAGMDzFoQcQDGIErLKBgFo2AUjAIMAAB7IE8zCM0gagAAAABJRU5ErkJggg==","orcid":"","institution":"Qilu Hospital of Shandong University","correspondingAuthor":true,"prefix":"","firstName":"Hui","middleName":"","lastName":"Tian","suffix":""}],"badges":[],"createdAt":"2025-07-17 16:38:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7151065/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7151065/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96653802,"identity":"2e6ffca9-77e1-45de-9966-cee2818e744c","added_by":"auto","created_at":"2025-11-24 16:38:48","extension":"tif","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2651842,"visible":true,"origin":"","legend":"","description":"","filename":"figure2.tif","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/a6c20bf810cc10b93489df45.tif"},{"id":96710739,"identity":"f34a9038-ad4b-4cb4-ac76-16723268202c","added_by":"auto","created_at":"2025-11-25 10:11:08","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1490011,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/813d3ac104bb84ff0c437725.docx"},{"id":96653801,"identity":"3390bfc1-7f84-4416-a9f8-a93d5e23b457","added_by":"auto","created_at":"2025-11-24 16:38:48","extension":"tif","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1598286,"visible":true,"origin":"","legend":"","description":"","filename":"figure3.tif","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/16d2a8f7f576706431fd2e62.tif"},{"id":96653770,"identity":"162b9d28-6b7e-4fc8-a4ac-e71330c665ac","added_by":"auto","created_at":"2025-11-24 16:38:46","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":37064,"visible":true,"origin":"","legend":"","description":"","filename":"Table1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/76a4f898ee7f86885055c5e7.docx"},{"id":96653807,"identity":"56094ff0-f7fc-4e71-a568-324a60d24754","added_by":"auto","created_at":"2025-11-24 16:38:48","extension":"tif","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1486706,"visible":true,"origin":"","legend":"","description":"","filename":"figure4.tif","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/b18d6d5ecd457024bd2c3026.tif"},{"id":96653781,"identity":"652e7c13-712c-4edb-93a6-08476c8740b0","added_by":"auto","created_at":"2025-11-24 16:38:47","extension":"tif","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1033386,"visible":true,"origin":"","legend":"","description":"","filename":"figure5.tif","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/f8a2ed6085f397afc659586a.tif"},{"id":96653813,"identity":"d02bc467-192d-438a-8421-49cfb9b05eca","added_by":"auto","created_at":"2025-11-24 16:38:49","extension":"tif","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1000726,"visible":true,"origin":"","legend":"","description":"","filename":"figure1.tif","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/be593cd49d56378f3a6e1763.tif"},{"id":96653773,"identity":"53aeb878-668b-4ba2-9097-856f3d7ac4d1","added_by":"auto","created_at":"2025-11-24 16:38:47","extension":"docx","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":30669,"visible":true,"origin":"","legend":"","description":"","filename":"Table2.docx","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/17278ce67337ea61a34dddda.docx"},{"id":96653815,"identity":"fe167e59-8a08-4bcf-b6d0-e3a2d5d2e79f","added_by":"auto","created_at":"2025-11-24 16:38:49","extension":"json","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10110,"visible":true,"origin":"","legend":"","description":"","filename":"cbedd097d7b840ef837b538c0c048912.json","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/89cf2c74de622b0a8b0b6afa.json"},{"id":96709218,"identity":"890c7594-fd4b-45d1-91c4-a928ed50090c","added_by":"auto","created_at":"2025-11-25 10:08:16","extension":"tif","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2541640,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile1.tif","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/eec9bdd34a9c40ddb1ae8004.tif"},{"id":96653783,"identity":"bd812d4d-c729-4cea-80eb-9d716ef9a167","added_by":"auto","created_at":"2025-11-24 16:38:47","extension":"xml","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":287638,"visible":true,"origin":"","legend":"","description":"","filename":"cbedd097d7b840ef837b538c0c0489121enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/b49a2b54c0441644ae08cb85.xml"},{"id":96709818,"identity":"ccf5abb7-48db-471f-8570-1ea11a097446","added_by":"auto","created_at":"2025-11-25 10:09:43","extension":"tif","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2651842,"visible":true,"origin":"","legend":"","description":"","filename":"figure2.tif","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/2381a59b1d48b04ae2ee8308.tif"},{"id":96653772,"identity":"771c51fb-eac7-42e0-8506-e526e3c8d71e","added_by":"auto","created_at":"2025-11-24 16:38:47","extension":"tif","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1598286,"visible":true,"origin":"","legend":"","description":"","filename":"figure3.tif","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/bc1de1b5d1f8b0afdcc7b865.tif"},{"id":96653787,"identity":"819514d6-8dab-4b04-87b2-67c3b8e25c1d","added_by":"auto","created_at":"2025-11-24 16:38:48","extension":"tif","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1486706,"visible":true,"origin":"","legend":"","description":"","filename":"figure4.tif","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/070c9e8613749748b35d62e6.tif"},{"id":96653797,"identity":"751a57d2-dea9-4eac-930e-2098e70f3fcf","added_by":"auto","created_at":"2025-11-24 16:38:48","extension":"tif","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1033386,"visible":true,"origin":"","legend":"","description":"","filename":"figure5.tif","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/bc2a8c20ec0809460605d5d2.tif"},{"id":96709752,"identity":"7dbf89e6-0752-456f-ab9e-1e568facce1d","added_by":"auto","created_at":"2025-11-25 10:09:36","extension":"tif","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1000726,"visible":true,"origin":"","legend":"","description":"","filename":"figure1.tif","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/f989312f727e93b12da215a3.tif"},{"id":96709640,"identity":"e0c7c4ac-a96a-484e-a12d-c5baabf7da1b","added_by":"auto","created_at":"2025-11-25 10:09:28","extension":"jpeg","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":323240,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/bb4fb356b70a2eae84b9e4b1.jpeg"},{"id":96653789,"identity":"d3d13821-5011-4b99-afd9-b5ab8544adae","added_by":"auto","created_at":"2025-11-24 16:38:48","extension":"jpeg","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":347516,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/f6502100504e93f1981a63d5.jpeg"},{"id":96653816,"identity":"483e60a3-09f0-4f29-9f44-593e619ee514","added_by":"auto","created_at":"2025-11-24 16:38:49","extension":"jpeg","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":213896,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/b089df052856553323c14d29.jpeg"},{"id":96709465,"identity":"72309c91-fd56-45bf-87f6-91087f05614c","added_by":"auto","created_at":"2025-11-25 10:09:03","extension":"jpeg","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":196918,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/4001d7f760d97bd09735ed17.jpeg"},{"id":96653809,"identity":"3847c0c2-171c-4d44-8718-8e16049a7698","added_by":"auto","created_at":"2025-11-24 16:38:49","extension":"jpeg","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":134730,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/5a15ae753116bf3421b043d2.jpeg"},{"id":96653782,"identity":"a5e55a71-8210-4046-b956-7a4f876ae629","added_by":"auto","created_at":"2025-11-24 16:38:47","extension":"jpeg","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":129344,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/6d16b3aa9206bfc52541a363.jpeg"},{"id":96653779,"identity":"85cdb42a-2c4d-443f-888f-e89b72ddb8d4","added_by":"auto","created_at":"2025-11-24 16:38:47","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":285573,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefigure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/404b6c2a774dc54d82bf5f94.png"},{"id":96653767,"identity":"aba20c29-ef03-4b65-89fa-cf16b77287ee","added_by":"auto","created_at":"2025-11-24 16:38:46","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":234110,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefigure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/c6a45f0e1a03fb97a483cc47.png"},{"id":96709213,"identity":"48ad7e88-5261-4c75-a692-5b1c3dfcf509","added_by":"auto","created_at":"2025-11-25 10:08:16","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":147501,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefigure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/29f692cbf313a95f4807a260.png"},{"id":96653777,"identity":"b02ed2a0-ae84-4cb3-8723-bf3b648b5d91","added_by":"auto","created_at":"2025-11-24 16:38:47","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":106639,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefigure5.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/bb0138f155c91e38499afe3c.png"},{"id":96710751,"identity":"a533b8f0-43b5-4c0c-b860-0b3fb5574800","added_by":"auto","created_at":"2025-11-25 10:11:09","extension":"png","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":67117,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefigure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/df2fa64a4c338543a7da0938.png"},{"id":96653800,"identity":"41d7c17f-65cb-4ec3-97f9-9b7cd32a8adc","added_by":"auto","created_at":"2025-11-24 16:38:48","extension":"png","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":31531,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/00d6ce9bac5a7f540e17e4ed.png"},{"id":96653786,"identity":"d5833ed7-2c58-438e-8353-0516c810e7bd","added_by":"auto","created_at":"2025-11-24 16:38:47","extension":"png","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56825,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/eed0bb7a788940998c74c926.png"},{"id":96653798,"identity":"9cc905a8-c38e-48cd-bc26-f742057e734c","added_by":"auto","created_at":"2025-11-24 16:38:48","extension":"png","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":43366,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/dcb1bee6af15d5bfd38ab4be.png"},{"id":96653765,"identity":"9acc039e-f14e-470f-bd7d-72449691d5ab","added_by":"auto","created_at":"2025-11-24 16:38:46","extension":"png","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":32468,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/1cabb9fc8f316b83b3d63c68.png"},{"id":96653784,"identity":"20df51c1-7472-4fa2-8ad0-679cb5efeae4","added_by":"auto","created_at":"2025-11-24 16:38:47","extension":"png","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":18195,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/ecaf7d63877736083a882738.png"},{"id":96653776,"identity":"bf608628-985b-43f2-980c-0c7de34ed194","added_by":"auto","created_at":"2025-11-24 16:38:47","extension":"png","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":26098,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/734f7a6057f8bfcc1c88028f.png"},{"id":96653812,"identity":"a122358f-83f3-460e-b206-1862f4462dd5","added_by":"auto","created_at":"2025-11-24 16:38:49","extension":"xml","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":287905,"visible":true,"origin":"","legend":"","description":"","filename":"cbedd097d7b840ef837b538c0c0489121structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/4b6e1ee86017878458aca1a8.xml"},{"id":96653778,"identity":"a877443b-09c7-4ecc-baa5-63d2593bdba1","added_by":"auto","created_at":"2025-11-24 16:38:47","extension":"html","order_by":34,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":298831,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/288c56b2337e51b59ca68ca0.html"},{"id":96653793,"identity":"d20a00c1-3071-4ddb-a3ba-bec01216bde5","added_by":"auto","created_at":"2025-11-24 16:38:48","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1915480,"visible":true,"origin":"","legend":"\u003cp\u003eFlowchart of patient selection and study design.\u003c/p\u003e","description":"","filename":"figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/c1fb167ebcd00450ea0e8cf1.png"},{"id":96653764,"identity":"dcbe3e80-6243-4169-a538-76ffce59d693","added_by":"auto","created_at":"2025-11-24 16:38:46","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":12630864,"visible":true,"origin":"","legend":"\u003cp\u003eROC curves at 1, 2, and 3 years and time-dependent AUC curves of eight machine learning models.\u003c/p\u003e","description":"","filename":"figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/2c5d73e7025f760d3d3b57ea.png"},{"id":96709715,"identity":"30c3529e-e35f-413f-9a2b-de04318ee0cc","added_by":"auto","created_at":"2025-11-25 10:09:33","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":7064802,"visible":true,"origin":"","legend":"\u003cp\u003eCalibration curves of the GBM model and decision curve analysis (DCA) of eight machine learning models at 1, 2, and 3 years.\u003c/p\u003e","description":"","filename":"figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/686c0f2dd72820bed391990b.png"},{"id":96653804,"identity":"d189a69d-c422-4908-a9b2-994271c34d17","added_by":"auto","created_at":"2025-11-24 16:38:48","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":4967445,"visible":true,"origin":"","legend":"\u003cp\u003eROC curves and calibration plots of the GBM model at 1, 2, and 3 years in the validation cohort.\u003c/p\u003e","description":"","filename":"figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/2da5b7b8c6317fb69c0ae91a.png"},{"id":96710021,"identity":"fdac14e3-98bf-4b5c-9266-d4919dd0e5aa","added_by":"auto","created_at":"2025-11-25 10:09:53","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":147589,"visible":true,"origin":"","legend":"\u003cp\u003eKaplan–Meier survival curves for high- and low-risk groups stratified by the GBM model in both the training and validation cohorts.\u003c/p\u003e","description":"","filename":"figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/9682854ba8982c7fb63d7962.png"},{"id":96913420,"identity":"68fdbd44-4961-4971-8b24-1e659eb13863","added_by":"auto","created_at":"2025-11-27 14:01:18","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":19659571,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/ade1e952-c59d-4f0a-b3ea-22d6a8d22323.pdf"},{"id":96653805,"identity":"4d03e69c-74ee-45ac-82bc-059fa9afe067","added_by":"auto","created_at":"2025-11-24 16:38:48","extension":"tif","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2541640,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAdditional file 1 \u003c/strong\u003eCalibration curves at 1-, 2-, and 3-year time points for seven machine learning models: RSF, LASSO-Cox, CoxBoost, SurvivalSVM, XGBoost, SuperPC, and plsRcox.\u003c/p\u003e","description":"","filename":"Additionalfile1.tif","url":"https://assets-eu.researchsquare.com/files/rs-7151065/v1/5187c167e036b5f4d651b867.tif"}],"financialInterests":"No competing interests reported.","formattedTitle":"Identifying High-Risk Features Associated with Limited Overall Survival in Esophageal Cancer Patients with Vascular Invasion Using Machine Learning","fulltext":[{"header":"Introduction","content":"\u003cp\u003eEsophageal cancer (EC) remains a significant global health burden, ranking as the seventh most commonly diagnosed cancer and the sixth leading cause of cancer-related mortality worldwide(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). According to GLOBOCAN 2020 data, an estimated 604,000 new EC cases and 544,000 related deaths were reported globally in that year(\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). Despite advancements in surgical techniques, chemotherapy, radiotherapy, and immunotherapy, the overall 5-year survival rate remains below 20%, largely due to the late-stage diagnosis and high recurrence rates(\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). Vascular invasion (VI), which includes both blood vessel invasion and lymphatic vessel invasion, is a critical pathological feature that significantly impacts disease progression(\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). Numerous studies have demonstrated that the presence of vascular invasion is associated with higher recurrence rates, increased distant metastasis, and reduced survival outcomes, particularly progression-free survival (PFS) and overall survival (OS)(\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAccurate identification of high-risk factors for shortened OS in patients with vascular invasion is essential for optimizing treatment strategies. Traditionally, prognosis is evaluated based on clinical staging systems such as the TNM classification, which considers tumor size, lymph node involvement, and distant metastasis(\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). However, vascular invasion is not explicitly included as a prognostic factor in the TNM system, despite its well-documented impact on recurrence and survival. This limitation is particularly relevant in early-stage esophageal cancer, where treatment decisions are often conservative when guided solely by TNM staging. In patients with positive vascular invasion, the biological behavior of the tumor may be far more aggressive than the staging suggests(\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). Vascular invasion is increasingly recognized as a marker of occult metastatic potential and is frequently associated with poorer clinical outcomes(\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). Xie et al. reported markedly reduced 1-, 3-, and 5-year survival rates in VI-positive compared to VI-negative patients (93.4% vs. 66.7%, 53.8% vs. 18.8%, and 48.1% vs. 15.6%, respectively)(\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). Even among patients classified as early stage, the presence of vascular invasion was independently associated with significantly worse overall survival(\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). These findings suggest that applying standard treatment protocols to VI-positive patients may lead to undertreatment in a subgroup with high risk of progression. Therefore, further risk stratification through the identification of additional prognostic indicators that contribute to poor prognosis in this patient population is warranted. Such efforts could inform decisions regarding the intensity of adjuvant therapy, frequency of follow-up, and need for more aggressive intervention, ultimately facilitating a more individualized and risk-adapted treatment approach.\u003c/p\u003e\u003cp\u003eIn recent years, machine learning (ML) has emerged as a powerful tool in oncologic research, offering improved predictive accuracy by integrating complex clinical, radiological, and pathological data(\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). ML models have been successfully applied to predict tumor recurrence, treatment response, and survival outcomes in various cancers, including lung, gastric, and colorectal cancers(\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). In esophageal cancer, ML-based approaches have been utilized for tumor classification, lymph node metastasis prediction, and response assessment to neoadjuvant therapy(\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). However, limited studies have focused on leveraging ML for predicting high-risk factors associated with shortened OS in esophageal cancer patients with vascular invasion. This study aims to develop and validate a machine learning-based model to identify high-risk features associated with limited OS in esophageal cancer patients with vascular invasion. By incorporating demographic, clinicopathologic, and radiological features, this model seeks to enhance risk stratification and improve preoperative decision-making. Identifying these prognostic factors can help refine treatment strategies, potentially guiding more aggressive surgical resections, adjuvant therapy recommendations, and closer postoperative surveillance for high-risk individuals.\u003c/p\u003e"},{"header":"Patients and methods","content":"\u003cp\u003e\u003cb\u003ePatients’ selection\u003c/b\u003e\u003c/p\u003e\u003cp\u003ewe systematically analyzed clinical data from a total of 354 patients who underwent surgical resection for esophageal cancer at Shandong University Qilu Hospital over a period spanning from January 2019 to December 2022. Patients were enrolled in this study based on the following criteria: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) Histologically confirmed diagnosis of esophageal squamous cell carcinoma or adenocarcinoma with positive vascular invasion confirmed by postoperative pathological examination; (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) Underwent curative-intent surgical resection, including esophagectomy combined with regional lymph node dissection; (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) Availability of complete clinicopathological data, including preoperative imaging, surgical records, and pathological reports; (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) Adequate follow-up data for progression-free survival analysis; (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e) No prior history of other malignancies that could affect vascular invasion status. Patients were excluded if they had a non-esophageal primary tumor, lacked evidence of vascular invasion, had incomplete clinical or pathological data, underwent non-curative surgery, lacked adequate follow-up information for overall survival analysis, or experienced perioperative mortality unrelated to tumor progression. Finally, we excluded 36 patients due to loss of follow-up. Based on strict adherence to the inclusion and exclusion criteria, a total of 318 patients were ultimately included in this study. This study received approval from the Medical Ethics Committee of Shandong University Qilu Hospital (approval number: KYLL-202008-023-1). Given the retrospective nature of this study utilizing previously collected clinical records, the institutional ethics committee granted a waiver of informed consent, and rigorous measures were implemented to protect patient confidentiality throughout the research process.\u003c/p\u003e\u003cp\u003e\u003cb\u003eData collection\u003c/b\u003e\u003c/p\u003e\u003cp\u003eWe collected a comprehensive range of clinical and pathological variables for all eligible patients. The dataset included: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) demographic characteristics such as sex, age, weight, body mass index (BMI), smoking and alcohol history, and comorbidities; (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) tumor-related features, including tumor location, length and adventitial invasion, as measured by preoperative imaging or endoscopy; (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) preoperative laboratory indicators including hematologic parameters, liver and renal function, glucose and lipid metabolism, and tumor markers; (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) treatment-related information including surgical approach, type of resection, and perioperative therapy; (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e) Pathological features such as histological type, tumor differentiation grade, T stage, N stage, and nerve invasion; (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e) follow-up data including survival status, survival time. To ensure data completeness and minimize potential bias, we applied a two-tiered approach. Variables with more than 30% missing data were excluded from further analysis to maintain data reliability. For variables with less than 30% missingness, missing values in continuous variables were imputed using the mean of the respective variable, while those in categorical variables were filled using the mode.\u003c/p\u003e\u003cp\u003e\u003cb\u003eTumor-related features\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTumor-related characteristics included tumor location, tumor length and Invasion of fibrous outer membrane. Tumor location was categorized as upper thoracic, middle thoracic, or lower thoracic esophagus, based on the anatomical segmentation defined by the 8th edition of the American Joint Committee on Cancer (AJCC) TNM staging system(\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). All patients underwent upper gastrointestinal radiography and contrast-enhanced computed tomography (CT) scans of the chest and upper abdomen within two weeks prior to surgery. The primary tumor site and longitudinal extent were primarily determined through gastrointestinal endoscopy. In cases where endoscopic documentation was incomplete or lacked specific measurements, tumor length was estimated using a combination of upper gastrointestinal contrast studies and contrast-enhanced-CT imaging. Invasion of the fibrous outer membrane was independently reviewed by two board-certified radiologists, each with over ten years of experience in thoracic oncology imaging. When imaging findings could not provide a definitive assessment, intraoperative observations were additionally considered for final documentation. Any discrepancies in interpretation were resolved by consensus discussion.\u003c/p\u003e\u003cp\u003e\u003cb\u003ePathological diagnosis\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe assessment of vascular invasion was performed on routinely processed postoperative specimens. All resected tissues were fixed in 10% neutral-buffered formalin, embedded in paraffin, and sectioned at 4-µm thickness. Hematoxylin and eosin (HE) staining was conducted in accordance with standardized histopathological protocols. Two experienced pathologists, blinded to the clinical characteristics and outcomes, independently evaluated each slide using light microscopy. Elastic or immunohistochemical staining was additionally employed when morphological assessment alone was insufficient to definitively identify vascular structures. Vascular invasion was defined as the unequivocal presence of tumor cells within the lumen of blood vessels or lymphatic channels, irrespective of the presence of an endothelial lining(\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e\u003ch2\u003eStatistical analysis\u003c/h2\u003e\u003cp\u003eStatistical analyses were conducted using R software (version 4.3.3). The analysis utilized several R packages, including randomForestSRC, CoxBoost, gbm, xgboost, glmnet, superpc, plsRcox and others. The 318 patients were randomized into training cohorts(TC) and validation cohorts(VC) in a 7:3 ratio using a random number seed. Continuous variables were summarized as mean ± standard deviation (SD) when normally distributed. Comparisons between groups were carried out using the Student’s t-test or Mann–Whitney U test, depending on data distribution. Categorical variables were described as counts and percentages, with group differences assessed using the chi-square test or Fisher’s exact test, as appropriate. OS was defined as the time from surgery to death from any cause or last follow-up. Kaplan–Meier survival curves were constructed to estimate OS, and statistical differences between subgroups were evaluated using the log-rank test. To explore prognostic factors, univariate Cox proportional hazards regression was first conducted. Variables demonstrating a P-value \u0026lt; 0.05 in univariate analysis were subsequently entered into a multivariate Cox model to identify independent predictors of limited OS. A two-sided P-value \u0026lt; 0.05 was considered to indicate statistical significance across all tests.\u003c/p\u003e\u003cp\u003e\u003cb\u003eBuilding Machine Learning Models\u003c/b\u003e\u003c/p\u003e\u003cp\u003eBased on feature selection and clinical relevance, eight key variables were incorporated into the modeling process, and eight machine learning algorithms were subsequently applied to analyze high-risk features and predict overall survival (OS) in patients with esophageal cancer and vascular invasion. These algorithms included Random Survival Forest (RSF), Gradient Boosting Machine (GBM), Least Absolute Shrinkage and Selection Operator Cox Regression (LASSO-Cox), Cox Proportional Hazards Model with Boosting (CoxBoost), Survival Support Vector Machine (survivalsvm), Extreme Gradient Boosting (XGBoost), Supervised Principal Components (SuperPC), Partial Least Squares Regression for Cox (plsRcox). Model performance was comprehensively evaluated using multiple metrics, including time-dependent area under the curve (Time-AUC), concordance index (C-index), receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis (DCA), across both training and validation cohorts. Time-AUC curves were generated to assess the dynamic predictive ability of each model at 1-, 2-, and 3-year time points, providing insights into their temporal discriminative power. Calibration plots were used to assess the agreement between predicted and observed outcomes, while DCA evaluated the net clinical benefit of each model across a range of threshold probabilities.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eBaseline Features of Patients\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA total of 318 esophageal cancer patients with vascular invasion from Qilu Hospital of Shandong University were included in this study, with 223 patients assigned to the training cohort and 95 to the validation cohort\u003cstrong\u003e\u0026nbsp;(Figure 1)\u003c/strong\u003e. Baseline characteristics are summarized in \u003cstrong\u003eTable 1\u003c/strong\u003e. Each cohort was further stratified into survival subgroups to facilitate comparison. In both cohorts, the mean age was approximately 63 years, and the male predominance was consistent\u0026nbsp;(TC: 81.8%\u0026ndash;86.7%, VC: 86.0%\u0026ndash;86.7%). Key demographic variables, including weight, BMI, smoking and alcohol history, hypertension, and diabetes, were comparable between alive and deceased groups within each cohort (\u003cem\u003eP\u003c/em\u003e \u0026gt; 0.05). Notable tumor-related differences were observed in both cohorts. Deceased patients had significantly larger tumors and a higher incidence of adventitial invasion compared to survivors (P \u0026lt; 0.01 in both TC and VC). Among laboratory indicators, higher white blood cell count (WBC), monocyte count, platelet count (PLT), serum sialic acid (SA), squamous cell carcinoma antigen (SCC), and serum ferritin (SF) levels were observed in deceased patients compared to survivors, with several of these variables showing significant differences. Albumin levels were significantly higher in survivors in the validation cohort (P = 0.041). Moreover, advanced T and N stage, as well as higher rates of nerve invasion (NI), were significantly associated with mortality in both cohorts (\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.05). To balance the two cohorts, baseline characteristics were also compared between the training and validation groups in \u003cstrong\u003etable1\u003c/strong\u003e. Except for differences in triglyceride levels, serum sialic acid, and the total number of dissected lymph nodes, all other variables showed no statistically significant differences between the two cohorts. Consequently, these findings support the use of the validation cohort for further assessment of model performance.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRisk Factor Selection and Model Development\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo further identify prognostic variables associated with limited OS, we performed univariate and multivariate Cox proportional hazards regression analyses. The results are summarized in \u003cstrong\u003etable 2\u003c/strong\u003e. In the univariate Cox regression analysis, eleven variables were found to be significantly associated with overall survival (OS), including BMI, tumor size, invasion of the fibrous outer membrane, WBC count, SA, SCC, SF, T stage, N stage, NI, and the total number of dissected lymph nodes (P \u0026lt; 0.05 for all). Variables with statistical significance in univariate analysis were subsequently included in the multivariate Cox regression model. The multivariate analysis identified the following eight variables as independent prognostic factors for limited OS: lower BMI (HR = 0.93, 95% CI: 0.87\u0026ndash;0.99, P = 0.019), invasion of the fibrous outer membrane (HR = 1.74, 95% CI: 1.14\u0026ndash;2.65, P = 0.011), elevated WBC count (HR = 1.19, 95% CI: 1.06\u0026ndash;1.34, P = 0.002), higher SCC levels (HR = 1.12, 95% CI: 1.04\u0026ndash;1.20, P = 0.002), advanced T stage (T3: HR = 4.65, 95% CI: 1.09\u0026ndash;19.87, P = 0.038; T4: HR = 6.31, 95% CI: 1.27\u0026ndash;31.28, P = 0.024), advanced N stage (N1: HR = 1.77, 95% CI: 1.00\u0026ndash;3.11, P = 0.049; N2: HR = 1.98, 95% CI: 1.08\u0026ndash;3.61, P = 0.027; N3: HR = 2.43, 95% CI: 1.25\u0026ndash;4.73, P = 0.009), presence of NI (HR = 1.54, 95% CI: 1.01\u0026ndash;2.34, P = 0.043), and total number of lymph nodes (HR = 1.03, 95% CI: 1.00\u0026ndash;1.05, P = 0.022). Using these selected variables, eight machine learning algorithms were developed to construct prognostic models: RSF, GBM, LASSO-Cox, CoxBoost, SurvivalSVM, XGBoost, SuperPC, and PLSR-Cox. Model performance was further systematically evaluated in both the training and validation cohorts to confirm predictive accuracy, model stability, and potential clinical relevance in survival risk stratification.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEvaluation of Prognostic Models Based on Eight Machine Learning\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe predictive performance of the eight machine learning\u0026ndash;based prognostic models was evaluated using time-dependent ROC curves at the 1-, 2-, and 3-year time points, as well as continuously measured area under the curve (AUC) values over the entire follow-up period. As shown in \u003cstrong\u003eFigure 2\u003c/strong\u003e, at 1 year, GBM achieved the highest AUC (0.987), followed closely by RSF (0.946) and XGBoost (0.945). At 2 and 3 years, GBM (AUC = 0.971 and 0.976, respectively) and RSF (AUC = 0.968 and 0.970) remained top performers. Specifically, GBM and RSF consistently demonstrated superior discriminative performance across all time points, with AUCs exceeding 0.94 at each interval. In contrast, models such as SurvivalSVM and SuperPC exhibited relatively lower discriminative ability, with AUCs ranging from 0.523 to 0.712 over the three time points. LASSO-Cox, CoxBoost, and PLSR-Cox models showed moderate performance, with AUC values generally between 0.75 and 0.85.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo further assess the agreement between predicted and observed survival probabilities, calibration curves were plotted for each model at the 1-, 2-, and 3-year time points. As shown in the calibration plots \u003cstrong\u003e(Figure 3a and Additional file 1)\u003c/strong\u003e, most models demonstrated acceptable alignment with the 45-degree reference line, indicating reasonable calibration performance. Among them, the GBM model exhibited the most consistent and closest fit to the ideal line across all three time points, reflecting excellent agreement between predicted and actual overall survival probabilities. Additionally, The RSF models also demonstrated good calibration, particularly at the 1- and 2-year marks, though slight deviations were noted at 3 years \u003cstrong\u003e(Additional file 1b)\u003c/strong\u003e. Moreover, to evaluate the potential clinical utility of each prognostic model, DCA was also conducted at the 1-, 2-, and 3-year time points \u003cstrong\u003e(Figure3)\u003c/strong\u003e. DCA estimates the net benefit of a prediction model across a range of threshold probabilities, thereby reflecting its value in assisting clinical decision-making. As shown in the DCA curves, the GBM model consistently yielded the highest net benefit across most threshold probabilities at all three time points. Although the RSF model exhibited relatively lower net benefit at the 1-year mark, its performance improved at the 2- and 3-year time points, showing moderate clinical utility in longer-term prognostic prediction. Taken together, considering its superior discrimination, excellent calibration, and the highest net clinical benefit across all time points, the GBM model was selected as the optimal prognostic model for overall survival prediction in this cohort.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInternal Validation\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;and Survival Stratification\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo further evaluate the generalizability of the GBM model, internal validation was performed using the independent test cohort \u003cstrong\u003e(Figure4)\u003c/strong\u003e. The model exhibited strong discriminatory capability, as evidenced by AUCs of 0.786, 0.774, and 0.750 at 1, 2, and 3 years, respectively, with consistent performance over time. Calibration plots also revealed a strong agreement between predicted and observed survival probabilities at 1, 2, and 3 years. These findings further support the robust performance and clinical applicability of the GBM model in prognostic risk estimation.\u003c/p\u003e\n\u003cp\u003eFurthermore, Kaplan\u0026ndash;Meier survival analysis was performed to evaluate the stratification ability of the GBM model \u003cstrong\u003e(Figure5)\u003c/strong\u003e. Based on the predicted risk scores, patients were divided into high-risk and low-risk groups using the median score as the cutoff. As shown in the survival curves, the high-risk group exhibited significantly worse OS compared to the low-risk group in both the training cohort (log-rank test, \u003cem\u003eP\u003c/em\u003e \u0026lt; 0.0001) and the validation cohort (\u003cem\u003eP\u003c/em\u003e = 0.0004). This clear separation between the two groups demonstrates the strong prognostic discriminatory power of the GBM model and its potential utility in individualized risk stratification.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn recent years, vascular invasion has gained increasing attention as a crucial pathological feature associated with poor prognosis and early recurrence in esophageal cancer. Defined as the presence of tumor cells within blood or lymphatic vessels, vascular invasion represents a direct route for tumor dissemination and metastasis. Multiple studies have demonstrated that the presence of vascular invasion correlates strongly with adverse clinical outcomes, including decreased PFS and OS\u0026nbsp;(19, 20). As such, patients with vascular invasion represent a clinically distinct subgroup that may benefit from intensified surveillance, enhanced postoperative monitoring, or escalated adjuvant treatment strategies. Despite its well-recognized prognostic significance, vascular invasion has not yet been incorporated into standard staging systems such as the TNM classification, which may lead to an underestimation of recurrence risk in affected patients. Our study aims to leverage machine learning techniques to identify high-risk factors associated with limited overall survival in esophageal cancer patients with vascular invasion, thereby enhancing precision in clinical decision-making and improving outcomes for this vulnerable patient population.\u003c/p\u003e\n\u003cp\u003eIn this study, we developed and validated multiple machine learning\u0026ndash;based prognostic models to predict overall survival in esophageal cancer patients with vascular invasion. By incorporating diverse clinical, pathological, imaging, and laboratory features, we successfully developed machine learning models to predict prognosis in esophageal cancer patients with vascular invasion. Compared to traditional survival modeling methods, such as Cox regression and traditional nomogram models, machine learning approaches allow for the integration of nonlinear interactions and high-dimensional feature spaces. In particular, we employed eight distinct algorithms spanning tree-based models (e.g., RSF, GBM, XGBoost), regularized regression models (e.g., LASSO-Cox, CoxBoost), and other survival learning frameworks (e.g., plsRcox, superPC, survivalSVM), enabling a robust comparative evaluation. Among all models, the GBM consistently outperformed others in both discrimination and calibration metrics, with the highest time-dependent AUCs at 1-, 2-, and 3-year follow-up points and strong net clinical benefit across decision curve analyses. GBM is a powerful ensemble learning algorithm that builds models sequentially, where each iteration focuses on minimizing the prediction error of the previous model(21). The GBM model\u0026rsquo;s superior performance can be attributed to its ability to iteratively refine predictions through boosting, effectively capturing subtle feature interactions and mitigating overfitting via embedded regularization technique(22). Survival curve stratification based on GBM risk scores also demonstrated clear prognostic separation, underscoring its clinical utility in individualized risk assessment.\u003c/p\u003e\n\u003cp\u003eThrough univariate and multivariate Cox regression analyses, eight independent prognostic factors were identified\u0026mdash;including fibrous outer membrane invasion, nerve invasion, T stage, N stage, BMI, WBC count, SCC level, and total number of lymph nodes dissected\u0026mdash;for inclusion in the final prognostic model. Among esophageal cancer patients with vascular invasion, both invasion of the fibrous outer membrane and nerve invasion were found to be significantly associated with poorer overall survival, consistent with previous studies that have identified these pathological features as independent adverse prognostic indicators(19, 23). This association may be attributed to several underlying mechanisms. The presence of nerve invasion and fibrous outer membrane invasion reflects deeper infiltration of tumor cells into surrounding anatomical structures. These features typically occur in more advanced disease stages and signify enhanced tumor aggressiveness and local invasiveness. Nerve invasion is often associated with perineural spread, allowing tumor cells to disseminate along nerve sheaths (24). Fibrous outer membrane invasion suggests that the tumor has breached the anatomical confines of the esophageal wall, increasing the likelihood of local recurrence or distant dissemination. Since this study focused on patients already exhibiting vascular invasion\u0026mdash;a known high-risk feature\u0026mdash;the coexistence of NI and outer membrane invasion further compounds the metastatic potential. This multifocal invasive behavior may indicate a biologically aggressive subtype prone to early recurrence and treatment resistance. Tumors infiltrating perineural or fibrous structures often reside in complex tissue interfaces, which may promote immune evasion and foster a more permissive microenvironment for progression(25). Such settings can impair treatment efficacy and contribute to poorer outcomes. Therefore, recognizing these features can support risk stratification and guide the implementation of more aggressive surveillance and therapeutic strategies. Additionally, both T and N stages were independently associated with overall survival(26). Specifically, advanced T stage reflects deeper tumor infiltration through the esophageal wall, which is often linked to greater tumor burden and increased likelihood of adjacent structure involvement (27). Similarly, higher N stage indicates more extensive regional lymph node metastasis, suggesting aggressive tumor biology and a higher risk of systemic dissemination(28). Taken together, these findings are also consistent with established oncologic principles and highlight the prognostic significance of tumor depth and nodal involvement even within this high-risk subgroup. Notably, in our study, only T stage \u0026ge; T3 showed a significant prognostic impact, underscoring that T2-stage tumors with vascular invasion may not independently confer limited OS, and suggesting that, while vascular invasion is an adverse feature, its prognostic effect may be modulated by the depth of tumor infiltration.\u003c/p\u003e\n\u003cp\u003eIn our study, lower BMI was significantly associated with poorer overall survival. This inverse relationship suggests that undernutrition or cancer-related cachexia may contribute to worse outcomes in this high-risk population. Malnutrition is known to impair immune function, delay postoperative recovery, and reduce tolerance to chemotherapy or radiotherapy(29, 30). Furthermore, a low BMI may reflect advanced disease burden and systemic inflammation, both of which are linked to unfavorable prognosis(31). These factors collectively compromise the patient\u0026apos;s physiological reserve and treatment responsiveness, ultimately contributing to poorer overall survival. Additionally, elevated WBC count was also associated with poorer overall survival in our study, suggesting that systemic inflammation may significantly contribute to tumor progression and adverse outcomes. Increased WBC levels reflect a heightened inflammatory state, which can promote tumor growth, angiogenesis, and metastasis through the release of cytokines and other mediators (32). In parallel, chronic inflammation may also induce oxidative stress and DNA damage, further accelerating tumor evolution. Moreover, inflammation-driven immune dysregulation may impair the body\u0026apos;s ability to mount an effective anti-tumor response, weakening host defense and facilitating disease progression (33). Thus, our findings and the above mechanisms indicate that elevated WBC is both a marker of systemic inflammation and a predictor of poor prognosis in this high-risk subgroup.\u003c/p\u003e\n\u003cp\u003eOur findings also indicated that SCC level was one of the key factors affecting overall survival in patients with vascular invasion. Mechanistically, SCC, a serological tumor marker secreted by squamous epithelial cells, has been associated with enhanced tumor cell proliferation, inhibition of apoptosis, epithelial-mesenchymal transition (EMT), and immune evasion(34). These biological processes contribute to increased tumor invasiveness and metastatic potential. Furthermore, in the context of vascular invasion, elevated SCC levels may indicate a more aggressive tumor phenotype characterized by increased angiogenesis, vascular permeability, and the potential for hematogenous dissemination. High SCC levels could reflect the extent of vascular involvement and the presence of micrometastases not captured by conventional imaging or staging systems. Therefore, SCC not only reflects disease severity but may also actively participate in tumor progression, justifying its prognostic value in patients with vascular invasion.\u003c/p\u003e\n\u003cp\u003eA notable finding in this analysis was that a higher total number of lymph nodes dissected was significantly associated with poorer overall survival in esophageal cancer patients with vascular invasion. While this finding may appear counterintuitive, it likely reflects the biological aggressiveness and extensive regional spread of tumors in this subgroup. A greater number of dissected lymph nodes often corresponds to more advanced disease with widespread microscopic metastases or pronounced lymphovascular infiltration, both of which portend a poor prognosis. Moreover, excessive lymphadenectomy may disrupt the local lymphatic architecture and immune surveillance, fostering a tumor-permissive microenvironment characterized by immune suppression and chronic inflammation(35). Such changes can impair the body\u0026rsquo;s antitumor immunity and promote residual tumor cell survival and dissemination. Additionally, extensive lymph node dissection may lead to increased postoperative complications and systemic inflammatory responses, further exacerbating patient outcomes(36). These findings underscore the importance of precise nodal staging and judicious lymphadenectomy. While adequate lymph node evaluation is essential for accurate prognosis and treatment planning, overtly aggressive dissection in biologically advanced disease may confer limited benefit and potential harm. Tailored surgical strategies that balance oncologic clearance with preservation of immune integrity are warranted in this high-risk population.\u003c/p\u003e\n\u003cp\u003eOverall, our study successfully established a robust and interpretable prognostic model using machine learning techniques, with the GBM model demonstrating strong discrimination, calibration, and clinical utility in esophageal cancer patients with vascular invasion. Despite the promising results, this study has several limitations that warrant consideration. First, the study was conducted retrospectively and based on data from a single institution, which may introduce selection bias and limit the generalizability of the findings. Although rigorous internal validation was performed, the robustness and applicability of the GBM model should be further confirmed through prospective validation using larger, independent, and multicenter cohorts that encompass broader demographic and clinical heterogeneity. Second, while machine learning algorithms such as GBM demonstrated superior predictive performance, one of the enduring criticisms of such models lies in their limited interpretability. Although GBM offers a better balance between accuracy and interpretability compared to more opaque models like deep neural networks, the clinical adoption of such tools depends on transparency and trust. Future work incorporating model explanation techniques, such as SHapley Additive exPlanations (SHAP), would be valuable in elucidating the contribution of individual features to survival predictions, thus enhancing interpretability and clinician confidence. Third, our model focused exclusively on OS as the primary outcome. While OS remains a fundamental endpoint in oncologic prognostication, additional endpoints such as DFS, PFS, and response to neoadjuvant or adjuvant therapies are equally critical in optimizing treatment strategies for esophageal cancer. Future studies should aim to develop and validate models tailored to these outcomes, potentially enabling more nuanced and comprehensive clinical decision-making. Lastly, this study did not incorporate certain potentially important variables\u0026mdash;such as molecular biomarkers, detailed nutritional assessments, and specific treatment parameters including radiotherapy dosage and chemotherapy protocols\u0026mdash;owing to limitations in data availability. Future studies that include these additional prognostic factors may contribute to the development of more comprehensive and accurate predictive models with broader clinical applicability.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study developed and validated a machine learning\u0026ndash;based prognostic model for esophageal cancer patients with vascular invasion, a subgroup known for poor outcomes. By integrating multiple clinical, pathological, and laboratory features, the GBM model demonstrated superior predictive performance in terms of discrimination, calibration, and clinical utility. The model effectively stratified patients into distinct risk groups, providing a reliable tool for individualized survival prediction. Importantly, several variables\u0026mdash;including nerve invasion, invasion of the fibrous outer membrane, T and N stage, BMI, WBC, SCC levels, and total number of dissected lymph nodes\u0026mdash;were identified as key predictors and offer mechanistic insights into disease progression. These findings underscore the potential of machine learning approaches to enhance risk assessment and guide treatment decision-making in esophageal cancer patients with vascular invasion. Future multicenter prospective studies and external validations are warranted to further confirm the generalizability and clinical applicability of this model.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eEC: Esophageal Cancer\u003c/p\u003e\n\u003cp\u003eVI: Vascular Invasion\u003c/p\u003e\n\u003cp\u003ePFS: Progression-Free Survival\u003c/p\u003e\n\u003cp\u003eDFS: Disease-Free Survival\u003c/p\u003e\n\u003cp\u003eOS: Overall Survival\u003c/p\u003e\n\u003cp\u003eML: Machine Learning\u003c/p\u003e\n\u003cp\u003eBMI: Body Mass Index\u003c/p\u003e\n\u003cp\u003eAJCC: American Joint Committee on Cancer\u003c/p\u003e\n\u003cp\u003eCT: Computed Tomography\u003c/p\u003e\n\u003cp\u003eHE: Hematoxylin and Eosin\u003c/p\u003e\n\u003cp\u003eWHO: World Health Organization\u003c/p\u003e\n\u003cp\u003eTC: Training Cohort\u003c/p\u003e\n\u003cp\u003eVC: Validation Cohort\u003c/p\u003e\n\u003cp\u003eSD: Standard Deviation\u003c/p\u003e\n\u003cp\u003eRSF: Random Survival Forest\u003c/p\u003e\n\u003cp\u003eGBM: Gradient Boosting Machine\u003c/p\u003e\n\u003cp\u003eLASSO-Cox: Least Absolute Shrinkage and Selection Operator Cox Regression\u003c/p\u003e\n\u003cp\u003eCoxBoost: Cox Proportional Hazards Model with Boosting\u003c/p\u003e\n\u003cp\u003eSurvivalsvm: Survival Support Vector Machine\u003c/p\u003e\n\u003cp\u003eXGBoost: Extreme Gradient Boosting\u003c/p\u003e\n\u003cp\u003eSuperPC: Supervised Principal Components\u003c/p\u003e\n\u003cp\u003ePlsRcox: Partial Least Squares Regression for Cox\u003c/p\u003e\n\u003cp\u003eTime-AUC: Time-dependent Area Under the Curve\u003c/p\u003e\n\u003cp\u003eAUC: Area Under the Curve\u003c/p\u003e\n\u003cp\u003eROC: Receiver Operating Characteristic\u003c/p\u003e\n\u003cp\u003eDCA: Decision Curve Analysis\u003c/p\u003e\n\u003cp\u003eWBC: White Blood Cell count\u003c/p\u003e\n\u003cp\u003ePLT: Platelet count\u003c/p\u003e\n\u003cp\u003eSA: Serum Sialic Acid\u003c/p\u003e\n\u003cp\u003eSCC: Squamous Cell Carcinoma antigen\u003c/p\u003e\n\u003cp\u003eSF: Serum Ferritin\u003c/p\u003e\n\u003cp\u003eNI: Nerve Invasion\u003c/p\u003e\n\u003cp\u003eHR: Hazard Ratio\u003c/p\u003e\n\u003cp\u003eCI: Confidence Interval\u003c/p\u003e\n\u003cp\u003eEMT: Epithelial-Mesenchymal Transition\u003c/p\u003e\n\u003cp\u003eSHAP: SHapley Additive exPlanations\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis retrospective study was approved by the Ethics Committee of Qilu Hospital, Shandong University (Approval number: KYLL-202008-023-1) and conducted in accordance with the ethical standards of the Declaration of Helsinki. Given the retrospective design and use of anonymized clinical data, the requirement for written informed consent was waived by the ethics committee.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for Publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis manuscript does not contain any identifiable personal data of individual participants in any form.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of Data and Materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAccess to the datasets analyzed in this study is available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThere are no conflicts of interest to disclose.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Taishan Scholar Program of Shandong Province (ts201712087) and the National Natural Science Foundation of China (Grant No. 82472814).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eYZ and JJL conceived and designed the study, performed data collection and statistical analysis, developed the machine learning models, and drafted the manuscript. ZZ, ZYL, \u0026nbsp;YJM, SJZ, and HML contributed to data acquisition, preprocessing, and result interpretation. JHQ assisted in model validation and manuscript formatting. HT supervised the project, provided methodological and conceptual guidance, and critically revised the manuscript. All authors read and approved the final version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe extend our sincere gratitude to the patients and medical staff of Qilu Hospital of Shandong University for their invaluable contributions to this study. Special thanks are also due to the radiologists and pathologists for their expert support in data interpretation and analysis.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eLi N, Chen S, Wang X, Zhang B, Zeng B, Sun C, et al. Identification of POU4F1 as a novel prognostic biomarker and therapeutic target in esophageal squamous cell carcinoma. Cancer Cell Int. 2024;24(1):280.\u003c/li\u003e\n\u003cli\u003eSheikh M, Roshandel G, McCormack V, Malekzadeh R. Current Status and Future Prospects for Esophageal Cancer. Cancers (Basel). 2023;15(3).\u003c/li\u003e\n\u003cli\u003eSung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, 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):209\u0026ndash;49.\u003c/li\u003e\n\u003cli\u003eMulisa G, Pero-Gascon R, McCormack V, Bisanz JE, Talukdar FR, Abebe T, et al. Multiple mycotoxin exposure assessment through human biomonitoring in an esophageal cancer case-control study in the Arsi-Bale districts of Oromia region of Ethiopia. Int J Hyg Environ Health. 2025;263:114466.\u003c/li\u003e\n\u003cli\u003eGonzalez J, Bahmad HF, Ocejo S, Abreu A, Popp M, Gogola S, et al. The Usefulness of Elastin Staining to Detect Vascular Invasion in Cancer. Int J Mol Sci. 2023;24(20).\u003c/li\u003e\n\u003cli\u003eOzer L, Senocak Tasci E, Mutlu AU, Piyade B, Ramoglu N, Ajredini M, et al. Intramural Component of Venous, Lymphatic, and Perineural Invasion in Colon Cancer: A Threat or an Illusion? Balkan Med J. 2022;39(6):436\u0026ndash;43.\u003c/li\u003e\n\u003cli\u003eLin Y, Zhang Y, Fang H, Hu Q, Duan H, Zhang L, et al. Survival and clinicopathological significance of blood vessel invasion in operable breast cancer: a systematic review and meta-analysis. Jpn J Clin Oncol. 2023;53(1):35\u0026ndash;45.\u003c/li\u003e\n\u003cli\u003eEngholm G, Lundberg FE, Konig SM, Olafsdottir E, Johannesen TB, Pettersson D, et al. TNM stage in the Nordic Cancer Registries 2004-2016: Registration and availability. Acta Oncol. 2024;63:303\u0026ndash;12.\u003c/li\u003e\n\u003cli\u003eWang A, Tan Y, Wang S, Chen X. The prognostic value of separate lymphatic invasion and vascular invasion in oesophageal squamous cell carcinoma: a meta-analysis and systematic review. BMC Cancer. 2022;22(1):1329.\u003c/li\u003e\n\u003cli\u003eCorachea AJM, Ferrer RJE, Ty LPB, Aquino LAC, Morta MT, Macalindong SS, et al. Lymphovascular Invasion Is Associated With Doxorubicin Resistance in Breast Cancer. Lab Invest. 2025;105(5):104115.\u003c/li\u003e\n\u003cli\u003eXie C, Chen Z, Xu J, Meng Z, Huang Z, Lin J. Influence of Lymphangio vascular (V) and perineural (N) invasion on survival of patients with resected esophageal squamous cell carcinoma (ESCC): a single-center retrospective study. PeerJ. 2022;10:e12974.\u003c/li\u003e\n\u003cli\u003eOguma J, Ishiyama K, Kurita D, Kanematsu K, Kubo K, Utsunomiya D, et al. Oncologic significance of lymphovascular invasion in patients with superficial esophageal squamous cell carcinoma reaching the muscularis mucosae or with slight invasion of the submucosa. Surgery. 2023;173(5):1162\u0026ndash;8.\u003c/li\u003e\n\u003cli\u003eSakkal M, Hajal AA. Machine learning predictions of tumor progression: How reliable are we? Comput Biol Med. 2025;191:110156.\u003c/li\u003e\n\u003cli\u003eFiste O, Liontos M, Zagouri F, Stamatakos G, Dimopoulos MA. Machine learning applications in gynecological cancer: A critical review. Crit Rev Oncol Hematol. 2022;179:103808.\u003c/li\u003e\n\u003cli\u003eHao Z, Wang Z, Ma J, Li Y, Zhang W. Harnessing the machine learning and nomogram models: elevating prognostication in nonmetastatic gastric cancer with \u0026quot;double invasion\u0026quot; for personalized patient care. Eur J Med Res. 2025;30(1):517.\u003c/li\u003e\n\u003cli\u003eMohammadian Rad N, Sosef O, Seegers J, Koolen L, Hoofwijk J, Woodruff HC, et al. Prognostic models for colorectal cancer recurrence using carcinoembryonic antigen measurements. Front Oncol. 2024;14:1368120.\u003c/li\u003e\n\u003cli\u003eHuang X, Wang Q, Xu W, Liu F, Pan L, Jiao H, et al. Machine learning to predict lymph node metastasis in T1 esophageal squamous cell carcinoma: a multicenter study. Int J Surg. 2024;110(12):7852\u0026ndash;9.\u003c/li\u003e\n\u003cli\u003eTeja R, Bhaskar R, Yadlapalli DC, Kukunuri S, Mohan R. Real-World Data of Esophageal Malignancies Managed Under the Government Scheme From a Tertiary Cancer Center in South India. Cureus. 2025;17(1):e77850.\u003c/li\u003e\n\u003cli\u003eZhou J, Yang Y, Zhang H, Luan S, Xiao X, Li X, et al. Lymphovascular and Perineural Invasion After Neoadjuvant Therapy in Esophageal Squamous Carcinoma. Ann Thorac Surg. 2023;115(6):1386\u0026ndash;94.\u003c/li\u003e\n\u003cli\u003eOguma J, Ishiyama K, Kurita D, Kanematsu K, Kubo K, Utsunomiya D, et al. Significance of lymphovascular invasion in esophageal squamous cell carcinoma undergoing neoadjuvant chemotherapy followed by esophagectomy. Esophagus. 2023;20(2):215\u0026ndash;24.\u003c/li\u003e\n\u003cli\u003eSalditt M, Humberg S, Nestler S. Gradient Tree Boosting for Hierarchical Data. Multivariate Behav Res. 2023;58(5):911\u0026ndash;37.\u003c/li\u003e\n\u003cli\u003eZhang X, Xue Y, Su X, Chen S, Liu K, Chen W, et al. A Transfer Learning Approach to Correct the Temporal Performance Drift of Clinical Prediction Models: Retrospective Cohort Study. JMIR Med Inform. 2022;10(11):e38053.\u003c/li\u003e\n\u003cli\u003eMatsumoto T, Noma K, Maeda N, Kato T, Moriwake K, Kawasaki K, et al. Safe and curative modified two-stage operation for T4 esophageal cancer after definitive chemoradiotherapy: a case report. Surg Case Rep. 2023;9(1):119.\u003c/li\u003e\n\u003cli\u003eZhu M, Luo F, Xu B, Xu J. Research Progress of Neural Invasion in Pancreatic Cancer. Curr Cancer Drug Targets. 2024;24(4):397\u0026ndash;410.\u003c/li\u003e\n\u003cli\u003eMa C, Luo H. A more novel and robust gene signature predicts outcome in patients with esophageal squamous cell carcinoma. Clin Res Hepatol Gastroenterol. 2022;46(10):102033.\u003c/li\u003e\n\u003cli\u003eKrauss DT, Schmidt T, Bruns CJ, Fuchs HF. [Evidence for the extent and oncological benefit of lymphadenectomy for esophageal cancer]. Chirurgie (Heidelb). 2025;96(4):273\u0026ndash;80.\u003c/li\u003e\n\u003cli\u003eYu Y, Wei X, Chen X, Li H, Liu Q, Sun H, et al. The T stage of esophageal cancer can be effectively predicted by muscularis propria thickness and muscularis propria + mucosa thickness under ultrasonic gastroscopy. Thorac Cancer. 2023;14(2):127\u0026ndash;34.\u003c/li\u003e\n\u003cli\u003eWang Z, Li F, Zhu M, Lu T, Wen L, Yang S, et al. Prognostic prediction and comparison of three staging programs for patients with advanced (T2-T4) esophageal squamous carcinoma after radical resection. Front Oncol. 2024;14:1376527.\u003c/li\u003e\n\u003cli\u003eNakayama K, Yoshida T, Nakayama Y, Iguchi N, Namba Y, Konishi M, et al. Activation of macrophages mediates dietary restriction-induced splenic involution. Life Sci. 2022;310:121068.\u003c/li\u003e\n\u003cli\u003eSantos SS, Costa L, Araripe TSO, Reges B, Ximenes HMA, Moreira A. Immunomodulatory enteral nutrition in post-surgical gastrointestinal cancer: Clinical, biochemical and nutritional impacts. Clin Nutr ESPEN. 2025;68:254\u0026ndash;62.\u003c/li\u003e\n\u003cli\u003eLocks LM, Parekh A, Newell K, Dauphinais MR, Cintron C, Maloomian K, et al. The ABCDs of Nutritional Assessment in Infectious Diseases Research. J Infect Dis. 2025;231(3):562\u0026ndash;72.\u003c/li\u003e\n\u003cli\u003eFruntelata RF, Bakri A, Stoica GA, Mogoanta L, Ionovici N, Popescu G, et al. Assessment of tumoral and peritumoral inflammatory reaction in cutaneous malignant melanomas. Rom J Morphol Embryol. 2023;64(1):41\u0026ndash;8.\u003c/li\u003e\n\u003cli\u003eJalali AM, Mitchell KJ, Pompoco C, Poludasu S, Tran S, Ramana KV. Therapeutic Significance of NLRP3 Inflammasome in Cancer: Friend or Foe? Int J Mol Sci. 2024;25(24).\u003c/li\u003e\n\u003cli\u003eHsu LC, Lin CN, Hsu FT, Chen YT, Chang PL, Hsieh LL, et al. Imipramine Suppresses Tumor Growth and Induces Apoptosis in Oral Squamous Cell Carcinoma: Targeting Multiple Processes and Signaling Pathways. Anticancer Res. 2023;43(9):3987\u0026ndash;96.\u003c/li\u003e\n\u003cli\u003ePatel AJ, Bille A. Lymph node dissection in lung cancer surgery. Front Surg. 2024;11:1389943.\u003c/li\u003e\n\u003cli\u003eBaud G, Jannin A, Marciniak C, Chevalier B, Do Cao C, Leteurtre E, et al. Impact of Lymph Node Dissection on Postoperative Complications of Total Thyroidectomy in Patients with Thyroid Carcinoma. Cancers (Basel). 2022;14(21).\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable 1.\u003c/strong\u003e Baseline characteristics of patients stratified by survival status in the training and validation cohorts.\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"936\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTC\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVC\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTC vs VC\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAlive (n=110)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDead (n=113)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ep value\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAlive (n=50)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDead (n=45)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ep value\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ep value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eAge (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e63.27 (8.13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e63.17 (7.89)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.922\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e63.54 (9.40)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e64.96 (8.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.437\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.327\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eGender (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.774\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e90 (81.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e98 (86.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e43 (86.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e39 (86.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eFemale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e20 (18.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e15 (13.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e7 (14.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e6 (13.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eWeight (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e66.42 (10.37)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e64.73 (10.85)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.235\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e64.18 (10.53)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e61.82 (9.73)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.261\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.053\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eBMI (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e24.33 (2.91)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e23.51 (3.34)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.051\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e23.47 (3.28)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e23.17 (2.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.126\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eSmoking history (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.263\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e40 (36.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e42 (37.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e26 (52.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e16 (35.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e70 (63.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e71 (62.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e24 (48.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e29 (64.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eHistory of alcohol intake (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.569\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.665\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.694\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e42 (38.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e38 (33.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e21 (42.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e16 (35.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e68 (61.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e75 (66.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e29 (58.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e29 (64.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eHypertension (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.168\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.215\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.958\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e85 (77.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e77 (68.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e40 (80.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e30 (66.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e25 (22.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e36 (31.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e10 (20.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e15 (33.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eDiabetes (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.936\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.279\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.984\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e94 (85.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e98 (86.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e45 (90.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e36 (80.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e16 (14.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e15 (13.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e5 (10.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e9 (20.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eTumor size (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e4.76 (1.85)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e5.51 (2.18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.006*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e4.68 (1.75)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e4.98 (2.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.478\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.189\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eTumor location (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.493\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.208\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eUpper esophagus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e5 (4.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e1 (0.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e3 (6.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e1 (2.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eMiddle esophagus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e20 (18.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e23 (20.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e8 (16.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e11 (24.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eLower esophagus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e53 (48.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e53 (46.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e28 (56.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e19 (42.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eUpper and middle esophagus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e3 (2.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e5 (4.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e3 (6.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e5 (11.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eMiddle and lower esophagus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e29 (26.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e31 (27.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e8 (16.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e9 (20.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.444\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eInvasion of fibrous outer membrane (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e\u0026lt;0.001*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026lt;0.001*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.812\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e85 (77.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e51 (45.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e41 (82.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e19 (42.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e25 (22.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e62 (54.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e9 (18.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e26 (57.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eHemoglobin (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e136.40 (16.16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e136.51 (18.20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.961\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e139.90 (17.23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e140.04 (15.79)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.966\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.092\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eWBC (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e5.73 (1.71)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e6.46 (1.60)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.001*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e5.77 (1.58)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e6.15 (1.77)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.265\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.466\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eNeutroCount (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e3.42 (1.41)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e4.55 (6.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.055\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e3.53 (1.46)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e3.82 (1.66)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.378\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.485\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eLymphoCount (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e1.71 (0.61)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e1.72 (0.55)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.855\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e1.66 (0.47)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e1.66 (0.43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.964\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.404\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eMonoCount (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e0.48 (0.22)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e0.54 (0.20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.033\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.43 (0.13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0.51 (0.18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.013*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.137\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003ePLT (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e232.12 (66.24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e248.69 (72.87)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.077\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e224.34 (58.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e252.71 (73.10)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.039*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.747\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eTotal protein (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e68.75 (5.09)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e68.44 (4.72)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.634\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e68.06 (5.13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e68.92 (9.15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.861\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eAlbumin (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e43.17 (3.27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e43.34 (3.43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.703\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e42.79 (3.39)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e44.13 (2.80)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.041*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.681\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003ePrealbumin (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e22.95 (4.63)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e22.88 (4.80)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.917\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e23.54 (3.94)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e23.68 (5.16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.881\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.223\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eTbil (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e11.52 (4.82)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e11.17 (4.75)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.587\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e12.12 (3.76)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e11.50 (3.72)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.417\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.379\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eDbil (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e6.89 (31.41)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e3.89 (1.50)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.312\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e3.94 (1.17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e4.07 (1.32)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.616\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.547\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eALT (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e14.82 (9.92)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e16.15 (12.16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.372\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e15.90 (6.65)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e17.31 (16.83)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.585\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.447\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eAST (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e18.56 (6.18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e18.96 (8.38)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.692\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e20.00 (6.84)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e19.67 (9.48)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.844\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.247\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eBUN (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e5.40 (1.99)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e5.39 (1.54)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.946\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e5.23 (1.32)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e5.48 (1.55)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.397\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.807\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eScr (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e75.54 (29.41)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e75.30 (15.35)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.938\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e74.58 (13.54)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e76.11 (16.12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.616\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.964\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eGlucose (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e6.17 (8.54)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e5.30 (1.09)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.288\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e5.55 (1.21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e5.69 (2.11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.691\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.862\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eCholesterol (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e4.75 (0.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e4.75 (1.06)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.979\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e4.77 (0.97)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e4.78 (0.96)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.978\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.842\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eHDL (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e1.25 (0.27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e1.25 (0.29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.938\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e1.25 (0.35)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e1.27 (0.26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.777\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.713\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eLDL (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e2.86 (0.70)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e2.92 (0.79)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.515\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e2.89 (0.65)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e2.85 (0.77)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.816\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.851\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eTriglyceride (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e1.18 (0.64)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e1.16 (0.51)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.783\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e1.34 (0.71)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e1.32 (0.66)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.862\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.032*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eUric acid (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e299.44 (60.50)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e302.15 (69.57)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.756\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e308.76 (80.38)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e303.13 (79.27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.732\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.537\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eLDH (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e193.28 (36.08)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e197.84 (33.25)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.327\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e192.22 (34.29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e196.07 (33.82)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.584\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.714\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eSA (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e59.17 (9.19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e61.70 (9.04)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.039*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e55.94 (8.08)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e60.09 (10.05)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.028*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.024*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eCEA (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e3.15 (3.90)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e3.29 (2.34)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e2.59 (1.73)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e2.98 (1.60)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.269\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.196\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eAFP (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e3.22 (1.55)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e2.84 (1.24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.046*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e3.07 (1.65)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e2.97 (1.21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.734\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.961\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eSCC (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e0.97 (0.64)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e1.61 (1.99)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.002*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e1.15 (1.05)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e1.55 (1.30)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.105\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.785\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eSF (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e218.18 (194.72)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e279.40 (169.45)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.013*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e227.93 (177.56)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e259.82 (160.24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.363\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.78\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eCA199 (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e16.73 (33.37)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e12.00 (13.96)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.167\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e12.70 (8.05)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e12.20 (8.49)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.773\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.485\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eCA125 (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e9.98 (5.51)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e9.32 (4.22)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.313\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e9.58 (4.16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e10.25 (4.40)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.452\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.664\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eCA724 (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e4.60 (9.81)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e3.51 (3.15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.264\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e3.92 (4.27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e2.71 (1.33)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.071\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.371\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003ePNI (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e51.70 (4.95)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e52.23 (5.55)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.455\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e51.65 (5.66)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e52.41 (3.98)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.456\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.955\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eSII (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e565.55 (616.91)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e648.78 (408.90)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.235\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e543.85 (382.70)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e626.49 (411.74)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.313\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eSurgical approach (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.367\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.469\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.947\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eMcKeown\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e49 (44.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e50 (44.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e22 (44.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e21 (46.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eIvor Lewis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e2 (1.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e6 (5.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e1 (2.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e3 (6.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eSweet\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e59 (53.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e57 (50.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e27 (54.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e21 (46.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eSurgical method (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.788\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.314\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.59\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eMinimally invasive surgery\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e46 (41.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e44 (38.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e21 (42.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e19 (42.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eRobotic surgery\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e4 (3.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e6 (5.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0 (0.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e2 (4.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eOpen surgery\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e60 (54.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e63 (55.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e29 (58.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e24 (53.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eNeoadjuvant therapy (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.938\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.824\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.631\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003e\u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e95 (86.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e99 (87.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e43 (86.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e37 (82.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003e\u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e15 (13.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e14 (12.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e7 (14.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e8 (17.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003ePathological type (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.326\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.694\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.562\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003e\u0026nbsp;ESCC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e88 (80.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e97 (85.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e42 (84.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e40 (88.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003e\u0026nbsp;EAC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e22 (20.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e16 (14.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e8 (16.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e5 (11.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eDifferentiation grade (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.117\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.053\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.92\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eWD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e7 (6.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e5 (4.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e6 (12.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0 (0.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eMD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e53 (48.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e41 (36.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e21 (42.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e20 (44.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eLD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e50 (45.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e67 (59.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e23 (46.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e25 (55.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eT stage (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e\u0026lt;0.001*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.023*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.455\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eT1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e19 (17.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e2 (1.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e8 (16.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e3 (6.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eT2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e36 (32.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e16 (14.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e19 (38.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e9 (20.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eT3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e53 (48.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e84 (74.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e23 (46.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e30 (66.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eT4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e2 (1.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e11 (9.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e0 (0.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e3 (6.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eN stage (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e\u0026lt;0.001*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.008*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.327\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eN0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e46 (41.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e20 (17.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e16 (32.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e10 (22.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eN1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e37 (33.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e37 (32.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e24 (48.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e11 (24.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eN2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e19 (17.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e33 (29.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e8 (16.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e19 (42.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eN3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e8 (7.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e23 (20.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e2 (4.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e5 (11.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eNI (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e\u0026lt;0.001*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026lt;0.001*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.309\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e91 (82.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e60 (53.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e41 (82.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e17 (37.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e19 (17.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e53 (46.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e9 (18.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e28 (62.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 274px;\"\u003e\n \u003cp\u003eTotal number of lymph nodes (mean (SD))\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e17.64 (7.93)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e20.26 (8.84)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 66px;\"\u003e\n \u003cp\u003e0.021*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e17.30 (7.97)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 104px;\"\u003e\n \u003cp\u003e16.44 (8.31)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76px;\"\u003e\n \u003cp\u003e0.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 85px;\"\u003e\n \u003cp\u003e0.045*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eAbbreviations: TC, Training Cohort; VC, Validation Cohort; BMI, Body Mass Index; WBC, White Blood Cell count; PLT, Platelet count; TBil, Total Bilirubin; DBil, Direct Bilirubin; ALT, Alanine Aminotransferase; AST, Aspartate Aminotransferase; BUN, Blood Urea Nitrogen; Scr, Serum Creatinine; HDL, High-Density Lipoprotein; LDL, Low-Density Lipoprotein; LDH, Lactate Dehydrogenase; SA, Serum Sialic Acid; CEA, Carcinoembryonic Antigen; AFP, Alpha-Fetoprotein; SCC, Squamous Cell Carcinoma antigen; SF, Serum Ferritin; CA199, Carbohydrate Antigen 199; CA125, Carbohydrate Antigen 125; CA724, Carbohydrate Antigen 724; PNI, Prognostic Nutritional Index; SII, Systemic Immune-Inflammation Index; ESCC, Esophageal Squamous Cell Carcinoma; EAC, Esophageal Adenocarcinoma; WD, Well Differentiated; MD, Moderately Differentiated; LD, Low Differentiation; NI, Nerve Invasion; SD, Standard Deviation.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable2\u0026nbsp;\u003c/strong\u003eUnivariate and multivariate Cox regression analyses identifying prognostic factors associated with overall survival.\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"740\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUnivariate analysis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMultivariate analysis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e \u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eOR (95%CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eP value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e \u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eOR (95%CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eP value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.00 (0.98-1.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.865\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eGender\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eFemale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.77 (0.44-1.32)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.337\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eWeight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.99 (0.97-1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.129\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eBMI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.93 (0.87-0.99)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.016*\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e0.93 (0.87-0.99)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e0.019*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eSmoking history\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.95 (0.65-1.39)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.782\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eHistory of alcohol intake\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.05 (0.71-1.56)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.791\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eHypertension\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.35 (0.91-2.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.138\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eDiabetes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003e\u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.94 (0.54-1.61)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.809\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eTumor size\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.14 (1.05-1.24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.002*\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e1.02 (0.92-1.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e0.682\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eTumor location\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eUpper esophagus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eMiddle esophagus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e3.64 (0.49-26.94)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.206\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eLower esophagus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e3.56 (0.49-25.78)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.208\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eUpper and middle esophagus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e5.10 (0.60-43.71)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.137\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eMiddle and lower esophagus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e3.80 (0.52-27.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.189\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eInvasion of fibrous outer membrane\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e2.66 (1.83-3.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026lt;0.001*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e1.74 (1.14-2.65)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e0.011*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eHemoglobin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.00 (0.99-1.01)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.917\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eWBC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.18 (1.07-1.30)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.001*\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e1.19 (1.06-1.34)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e0.002*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eNeutroCount\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.02 (1.00-1.05)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.089\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eLymphoCount\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.01 (0.74-1.38)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.994\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eMonoCount\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.91 (0.99-3.66)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.053\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003ePLT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.00 (1.00-1.01)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.061\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eTotal protein\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.99 (0.95-1.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.505\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eAlbumin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.00 (0.95-1.06)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.960\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003ePrealbumin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.99 (0.95-1.03)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.638\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eTbil\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.99 (0.95-1.03)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.589\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eDbil\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.99 (0.92-1.05)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.672\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eALT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.00 (0.99-1.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.692\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eAST\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.00 (0.97-1.03)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.962\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eBUN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.00 (0.91-1.11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.974\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eScr\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.00 (0.99-1.01)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.992\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eGlucose\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.97 (0.88-1.06)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.491\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eCholesterol\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.99 (0.81-1.21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.902\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eHDL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.04 (0.54-2.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.903\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eLDL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.06 (0.82-1.38)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.653\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eTriglyceride\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.93 (0.66-1.31)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.682\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eUric acid\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.00 (1.00-1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.920\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eLDH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.00 (1.00-1.01)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.246\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eSA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.02 (1.00-1.04)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.028*\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e1.00 (0.98-1.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e0.856\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eCEA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.00 (0.96-1.06)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.848\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eAFP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.87 (0.75-1.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.081\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eSCC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.11 (1.04-1.18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026lt;0.001*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e1.12 (1.04-1.20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e0.002*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eSF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.00 (1.00-1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.025*\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e1.00 (1.00-1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e0.264\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eCA199\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.99 (0.98-1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.208\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eCA125\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.98 (0.95-1.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.440\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eCA724\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.98 (0.95-1.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.368\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003ePNI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.01 (0.97-1.05)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.582\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eSII\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.00 (1.00-1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.202\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eSurgical approach\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eMcKeown\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eIvor Lewis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.73 (0.74-4.04)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.204\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eSweet\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.97 (0.66-1.41)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.863\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eSurgical method\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eMinimally invasive surgery\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eRobotic surgery\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.42 (0.61-3.34)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.416\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eOpen surgery\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.07 (0.73-1.58)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.717\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eNeoadjuvant therapy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.87 (0.49-1.52)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.615\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003ePathological type\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eESCC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eEAC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e0.76 (0.45-1.28)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.300\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eDifferentiation grade\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eWD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eMD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.10 (0.44-2.79)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.836\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eLD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.62 (0.65-4.03)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.296\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eT stage\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eT1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eT2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e3.94 (0.91-17.15)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.067\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e2.80 (0.63-12.42)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e0.175\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eT3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e9.40 (2.31-38.24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.002*\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e4.65 (1.09-19.87)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e0.038*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eT4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e24.11 (5.33-109.09)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026lt;0.001*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e6.31 (1.27-31.28)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e0.024*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eN stage\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eN0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eN1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.99 (1.16-3.43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.013*\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e1.77 (1.00-3.11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e0.049*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eN2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e3.01 (1.72-5.25)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026lt;0.001*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e1.98 (1.08-3.61)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e0.027*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eN3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e4.27 (2.34-7.80)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026lt;0.001*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e2.43 (1.25-4.73)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e0.009*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eNI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003eRef\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e2.49 (1.72-3.61)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e\u0026lt;0.001*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e1.54 (1.01-2.34)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e0.043*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 261px;\"\u003e\n \u003cp\u003eTotal number of lymph nodes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 154px;\"\u003e\n \u003cp\u003e1.03 (1.01-1.06)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 79px;\"\u003e\n \u003cp\u003e0.002*\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 16px;\"\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 160px;\"\u003e\n \u003cp\u003e1.03 (1.00-1.05)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 69px;\"\u003e\n \u003cp\u003e0.022*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eAbbreviations: OR, Odds Ratio; Ref, Reference; BMI, Body Mass Index; WBC, White Blood Cell count; PLT, Platelet count; TBil, Total Bilirubin; DBil, Direct Bilirubin; ALT, Alanine Aminotransferase; AST, Aspartate Aminotransferase; BUN, Blood Urea Nitrogen; Scr, Serum Creatinine; HDL, High-Density Lipoprotein; LDL, Low-Density Lipoprotein; LDH, Lactate Dehydrogenase; SA, Serum Sialic Acid; CEA, Carcinoembryonic Antigen; AFP, Alpha-Fetoprotein; SCC, Squamous Cell Carcinoma antigen; SF, Serum Ferritin; CA199, Carbohydrate Antigen 199; CA125, Carbohydrate Antigen 125; CA724, Carbohydrate Antigen 724; PNI, Prognostic Nutritional Index; SII, Systemic Immune-Inflammation Index; ESCC, Esophageal Squamous Cell Carcinoma; EAC, Esophageal Adenocarcinoma; WD, Well Differentiated; MD, Moderately Differentiated; LD, Low Differentiation; NI, Nerve Invasion;\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"esophageal cancer, vascular invasion, prognostic model, machine learning, survival prediction, risk stratification","lastPublishedDoi":"10.21203/rs.3.rs-7151065/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7151065/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eVascular invasion is a critical pathological feature associated with poor prognosis in esophageal cancer, yet individualized risk prediction in this subgroup remains limited.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eWe retrospectively analyzed clinical data from 318 esophageal cancer patients with confirmed vascular invasion who underwent surgical resection at Shandong University Qilu Hospital between January 2019 and December 2022. Eight machine learning models were constructed using clinical, pathological, and laboratory features. The Gradient Boosting Machine (GBM) model was selected based on superior performance in discrimination, calibration, and decision curve analysis. Internal validation and survival stratification were conducted to assess robustness and clinical utility.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eEight variables were identified as independent prognostic factors: nerve invasion, invasion of the fibrous outer membrane, T stage, N stage, BMI, white blood cell count, squamous cell carcinoma antigen level, and total number of lymph nodes dissected. The GBM model achieved the highest time-dependent AUCs (1-year: 0.987, 2-year: 0.971, 3-year: 0.976) and demonstrated consistent calibration and net clinical benefit. Survival analysis based on GBM risk scores revealed significant stratification between risk groups (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001).\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e\u003cp\u003eOur GBM-based model provides accurate and interpretable prognostic predictions for esophageal cancer patients with vascular invasion. It offers valuable guidance for individualized clinical decision-making and warrants further validation in multicenter prospective cohorts.\u003c/p\u003e","manuscriptTitle":"Identifying High-Risk Features Associated with Limited Overall Survival in Esophageal Cancer Patients with Vascular Invasion Using Machine Learning","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-24 16:38:36","doi":"10.21203/rs.3.rs-7151065/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2025-12-02T03:46:15+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"148211625375163777995387197306787787158","date":"2025-11-27T02:38:22+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-11-12T12:58:53+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-07-23T07:56:40+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-21T05:54:19+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-18T12:26:48+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-07-17T16:23:28+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"59f249df-c236-42c5-9455-546fe9f603e7","owner":[],"postedDate":"November 24th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":58434813,"name":"Biological sciences/Cancer"},{"id":58434814,"name":"Health sciences/Oncology"}],"tags":[],"updatedAt":"2025-11-24T16:38:36+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-24 16:38:36","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7151065","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7151065","identity":"rs-7151065","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}