Xixun Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Xixun Wang has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
- Lixin Jiang
Department of Gastrointestinal Surgery, Yantai Yuhuangding Hospital Affiliated to Medical College of Qingdao University 1 , Yantai 264000, Shandong
2 shared publications - Mi Jian
Department of Gastrointestinal Surgery, Yantai Yuhuangding Hospital, Qingdao University, Yantai, China
2 shared publications - Miaomiao Mm Li
Department of Gastrointestinal Surgery, Yantai Yuhuangding Hospital, Qingdao University, Yantai, China
2 shared publications - Ruyue Mm Chen
Department of Gastrointestinal Surgery, Yantai Yuhuangding Hospital, Qingdao University, Yantai, China
2 shared publications - Yifei Zhang
Department of Gastrointestinal Surgery, Yantai Yuhuangding Hospital, Qingdao University, Yantai, China
2 shared publications - Aina Liu
Department of Oncology, The Affiliated Yantai Yuhuangding Hospital of Qingdao University, Yantai, China
1 shared publication
Latest Publications
- Open Access
- 19 August 2026
- Published by Springer Nature in Cancer Immunology, Immunotherapy
- Vol. 75 (11)
Interpretable machine learning for predicting immunotherapy response in gastric cancer via serum cytokine profiling and organoid co-culture platforms
Background Immunotherapy has become a promising strategy for gastric cancer, yet predicting treatment response remains challenging. We developed an exploratory, biologically informed machine learning framework integrating peri-treatment serum biomarkers and clinical variables, with candidate cytokines informed by an in vitro patient-derived organoids (PDOs)–cancer-associated fibroblasts (CAFs)–T cells co-culture platform, to assess pathological response to PD-1 inhibitor (PD-1i)-based therapy in gastric cancer. Methods Data from 125 patients receiving PD-1i were analyzed. Serum samples collected before and after immunotherapy were tested for cytokines. Eighteen successfully established PDOs-CAFs-T cells co-culture models were used for cytokine profiling and to identify seven candidate cytokines. These cytokines were subsequently measured in paired serum samples from the 125-patient clinical cohort and combined with clinicopathological variables for machine learning model development. Tumor Regression Grade (TRG) served as the binary outcome variable, representing patients with good versus poor/no tumor regression. Experimental and clinical data were used to train predictive models with multiple algorithms, including CatBoost and several other machine learning methods. Model performance was assessed by tenfold cross-validation, ROC, precision-recall (PR) Curve, and decision curve analyses. Feature importance was interpreted using Shapley additive explanations (SHAP) and visualized using decision tree. Results The PDOs-CAFs-T cells co-culture platform showed cytokine-change trends that were generally concordant with those observed in patient serum after treatment. 10 key predictors were identified, including pre-/post-treatment maximum diameter, ycT/ycN stages, and pre- or post-treatment cytokines (IFN-α2, IL-2, IL-6, IL-8, MCP-1). In the internal testing set, the tuned CatBoost model achieved an AUC of 0.845. SHAP analysis revealed post-treatment maximum diameter and cytokine profiles as the most influential features. Decision trees clearly and interpretably present the model. Conclusion This study presents an exploratory, biologically informed machine learning framework for peri-treatment assessment of pathological response in gastric cancer. The PDOs-CAFs-T cells co-culture platform provided biological context for candidate cytokine selection.
- 19 August 2026
- Published by Springer Nature in Cancer Immunology, Immunotherapy
- Vol. 75 (11)
Open AccessInterpretable machine learning for predicting immunotherapy response in gastric cancer via serum cytokine profiling and organoid co-culture platforms
Background Immunotherapy has become a promising strategy for gastric cancer, yet predicting treatment response remains challenging. We developed an exploratory, biologically informed machine learning framework integrating peri-treatment serum biomarkers and clinical variables, with candidate cytokines informed by an in vitro patient-derived organoids (PDOs)–cancer-associated fibroblasts (CAFs)–T cells co-culture platform, to assess pathological response to PD-1 inhibitor (PD-1i)-based therapy in gastric cancer. Methods Data from 125 patients receiving PD-1i were analyzed. Serum samples collected before and after immunotherapy were tested for cytokines. Eighteen successfully established PDOs-CAFs-T cells co-culture models were used for cytokine profiling and to identify seven candidate cytokines. These cytokines were subsequently measured in paired serum samples from the 125-patient clinical cohort and combined with clinicopathological variables for machine learning model development. Tumor Regression Grade (TRG) served as the binary outcome variable, representing patients with good versus poor/no tumor regression. Experimental and clinical data were used to train predictive models with multiple algorithms, including CatBoost and several other machine learning methods. Model performance was assessed by tenfold cross-validation, ROC, precision-recall (PR) Curve, and decision curve analyses. Feature importance was interpreted using Shapley additive explanations (SHAP) and visualized using decision tree. Results The PDOs-CAFs-T cells co-culture platform showed cytokine-change trends that were generally concordant with those observed in patient serum after treatment. 10 key predictors were identified, including pre-/post-treatment maximum diameter, ycT/ycN stages, and pre- or post-treatment cytokines (IFN-α2, IL-2, IL-6, IL-8, MCP-1). In the internal testing set, the tuned CatBoost model achieved an AUC of 0.845. SHAP analysis revealed post-treatment maximum diameter and cytokine profiles as the most influential features. Decision trees clearly and interpretably present the model. Conclusion This study presents an exploratory, biologically informed machine learning framework for peri-treatment assessment of pathological response in gastric cancer. The PDOs-CAFs-T cells co-culture platform provided biological context for candidate cytokine selection.
- Open Access
- 4 October 2024
- Published by Wolters Kluwer Health in Medicine
- Vol. 103 (40) , e39910
IGFL2 expression and surgical volume: Independent predictors of survival in gastric cancer
This study aimed to assess the impact of surgeons’ annual volume and insulin-like growth factor-like family member 2 (IGFL2) expression on gastric cancer prognosis. Clinicopathological data from 475 patients who underwent D2 lymph node dissection were analyzed. IGFL2 expression was evaluated using immunohistochemistry. Patients were divided into training (70%) and validation (30%) groups. Univariate and multivariate Cox regression identified risk factors for overall survival (OS) and disease-free survival (DFS), leading to a clinical prediction model. Model performance was evaluated using C-index. High IGFL2 expression and low surgical volume independently predicted poorer OS and DFS (hazard ratio = 2.13, 2.17, all P < .01). Surgeons performing >26 cases annually had higher OS and DFS (hazard ratio = 1.65, 1.58, all P < .01). Nomograms integrating surgical volume, IGFL2 expression, grade, TNM staging, and carcinoembryonic antigen showed superior predictive accuracy for OS and DFS compared to TNM alone, with robust C-indices and area under the curve values. Surgeons’ annual volume and IGFL2 expression independently predict gastric cancer prognosis, emphasizing the need for specialized training and further research on IGFL2’s molecular mechanisms to enhance patient outcomes.
- 4 October 2024
- Published by Wolters Kluwer Health in Medicine
- Vol. 103 (40) , e39910
Open AccessIGFL2 expression and surgical volume: Independent predictors of survival in gastric cancer
This study aimed to assess the impact of surgeons’ annual volume and insulin-like growth factor-like family member 2 (IGFL2) expression on gastric cancer prognosis. Clinicopathological data from 475 patients who underwent D2 lymph node dissection were analyzed. IGFL2 expression was evaluated using immunohistochemistry. Patients were divided into training (70%) and validation (30%) groups. Univariate and multivariate Cox regression identified risk factors for overall survival (OS) and disease-free survival (DFS), leading to a clinical prediction model. Model performance was evaluated using C-index. High IGFL2 expression and low surgical volume independently predicted poorer OS and DFS (hazard ratio = 2.13, 2.17, all P < .01). Surgeons performing >26 cases annually had higher OS and DFS (hazard ratio = 1.65, 1.58, all P < .01). Nomograms integrating surgical volume, IGFL2 expression, grade, TNM staging, and carcinoembryonic antigen showed superior predictive accuracy for OS and DFS compared to TNM alone, with robust C-indices and area under the curve values. Surgeons’ annual volume and IGFL2 expression independently predict gastric cancer prognosis, emphasizing the need for specialized training and further research on IGFL2’s molecular mechanisms to enhance patient outcomes.
- Open Access
- 4 January 2023
- Published by Frontiers Media SA in Frontiers in Surgery
- Vol. 9, 1003525
Risk factor analysis and construction of prediction models for short-term postoperative complications in patients undergoing gastrointestinal tract surgery
Purpose: To identify risk factors associated with short-term postoperative complications in patients with gastrointestinal cancer and develop and validate prediction models to predict the probability of complications.Methods: A total of 335 patients enrolled in the primary cohort of this study were divided into training and validation sets in a chronological order. Using univariate and multivariate logistic regression analyses, the risk factors for postoperative complications were determined, and nomogram prediction models were constructed. The performance of the nomogram was assessed with respect to the receiver operator characteristic and calibration curves.Results: Patients with complications had a stronger postoperative stress response and a longer duration of daily fluid intake/output ratio >1 after surgery. Logistic analysis revealed that body mass index (BMI), body temperature on POD4 (T.POD4), neutrophil percentage on POD4 (N.POD4), fasting blood glucose on POD4 (FBG.POD4), and the presence of fluid intake/output ratio 1 were risk factors for POD30 complications. The areas under the curve of Nomogram-A for POD7 complications were 0.867 and 0.833 and those of Nomogram-B for POD30 complications were 0.920 and 0.918 in the primary and validation cohorts, respectively. The calibration curves showed good consistency in both cohorts.Conclusion: This study presented two nomogram models to predict short-term postoperative complications in patients with gastrointestinal cancer. The results could help clinicians identify patients at high risk of complications within POD7 or POD30.
- 4 January 2023
- Published by Frontiers Media SA in Frontiers in Surgery
- Vol. 9, 1003525
Open AccessRisk factor analysis and construction of prediction models for short-term postoperative complications in patients undergoing gastrointestinal tract surgery
Purpose: To identify risk factors associated with short-term postoperative complications in patients with gastrointestinal cancer and develop and validate prediction models to predict the probability of complications.Methods: A total of 335 patients enrolled in the primary cohort of this study were divided into training and validation sets in a chronological order. Using univariate and multivariate logistic regression analyses, the risk factors for postoperative complications were determined, and nomogram prediction models were constructed. The performance of the nomogram was assessed with respect to the receiver operator characteristic and calibration curves.Results: Patients with complications had a stronger postoperative stress response and a longer duration of daily fluid intake/output ratio >1 after surgery. Logistic analysis revealed that body mass index (BMI), body temperature on POD4 (T.POD4), neutrophil percentage on POD4 (N.POD4), fasting blood glucose on POD4 (FBG.POD4), and the presence of fluid intake/output ratio 1 were risk factors for POD30 complications. The areas under the curve of Nomogram-A for POD7 complications were 0.867 and 0.833 and those of Nomogram-B for POD30 complications were 0.920 and 0.918 in the primary and validation cohorts, respectively. The calibration curves showed good consistency in both cohorts.Conclusion: This study presented two nomogram models to predict short-term postoperative complications in patients with gastrointestinal cancer. The results could help clinicians identify patients at high risk of complications within POD7 or POD30.