Proposed Models for Prediction of Mortality in Stage-I and Stage-II Gastric Cancer and 5 Years after Radical Gastrectomy

Predictive Mortality and Gastric Cancer Risk Using Clinical and Socio-Economic Data: A Nationwide Multicenter Cohort Study

Background/Objectives: Gastric cancer is a leading cause of cancer-related mortality, particularly in East Asia, with a notable burden in Republic of Korea. This study aimed to construct and develop machine learning models for the prediction of gastric cancer mortality and the identification of risk factors. Methods: All data were acquired from the Korean Clinical Data Utilization for Research Excellence by multiple medical centers in South Korea. A total of 23,717 gastric cancer patients were divided into two groups by cause of mortality (all-cause of 2664 and disease-specific of 1620) and investigated. We used comprehensive data integrating clinical, pathological, lifestyle, and socio-economic factors. Cox proportional hazards analysis was conducted to estimate hazard ratios for mortality. Five machine learning models (random forest, gradient boosting machine, XGBoost, light GBM, and cat boosting) were developed to predict mortality. The models were interpreted by SHAP, one of the explainable AI techniques. Results: For all-cause mortality, the gradient-boosting machine learning model demonstrated the highest performance with an AUC-ROC of 0.795. For disease-specific mortality, the light GBM model outperformed others, achieving an AUC-ROC of 0.867. Significant predictors included the AJCC7 stage, tumor size, lymph node count, and lifestyle factors such as smoking, drinking, and diabetes. Conclusions: This study underscores the importance of integrating both clinical and lifestyle data to enhance mortality prediction accuracy in gastric cancer patients. The findings highlight the need for personalized treatment approaches in the Korean population and emphasize the role of demographic-specific data in predictive modeling.

Systemic immune inflammation index and gastric cancer prognosis: A systematic review and meta‑analysis

The present study aimed to pool the available data on the associations between the systemic immune inflammation index (SII) and overall survival (OS) or recurrence-free survival (RFS) in patients with gastric cancer (GC). A systematic search was conducted in the PubMed, EMBASE and Scopus databases for observational studies, and a random effects model was used to conduct the statistical analysis. Pooled effect sizes were reported as hazard ratios (HRs) with corresponding 95% confidence intervals (CI). Data from 30 studies (24 conducted in China) with follow-ups ranging between 15.5 and 65.6 months were analyzed. Patients with GC and high SII levels had poor OS (HR, 1.53; 95% CI, 1.34-1.75) and recurrence free survival (HR, 1.41; 95% CI, 1.17-1.70). These increased risks were present irrespective of the treatment strategy (surgical or non-surgical management), the sample size (<500 and ≥500) and the cut-off used to define high and low SII (<600 and ≥600 x109 cells/l). The results of this meta-analysis suggest that high pretreatment SII levels were associated with poor OS and RFS in patients with GC.

Identification and validation of molecular subtypes and prognostic signature for stage I and stage II gastric cancer based on neutrophil extracellular traps

Purpose

This study identified subtypes and prognostic signature of stage I and stage II gastric cancer based on neutrophil extracellular trap (NET)-related genes.

Methods

The gene expression data associated with stage I and stage II gastric cancer were downloaded from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. NET-related genes were obtained from previous reference. Differentially expressed NET-related genes were selected by consensus cluster analysis. The differences in immune infiltration between two subtypes were analyzed. Prognosis-related genes were further screened by univariate Cox regression analysis. Gene Set Enrichment Analysis (GSEA) of prognostic signatures was conducted with clusterprofiler. Finally, a miRNA-mRNA-transcription factor (TF) network was constructed.

Results

Total 43 differential NET-related genes were obtained and two subtypes were obtained based on these genes. Patients of cluster 2 had a better prognosis compared to cluster 1. Eight types of immune cells were differential in infiltration level between two subtypes. Following univariate Cox regression analysis, two genes of CXC chemokine receptor 4 (CXCR4) and nuclear factor, erythroid 2-like 2 (NFE2L2) significantly related to patient survival were selected. GSEA of single gene revealed that CXCR4 was associated with allograft rejection and NFE2L2 was associated with drug metabolism-cytochrome P450. A network with 421 miRNA-mRNA-TF regulatory pairs was constructed.

Conclusion

The present study identified two subtypes and a prognostic signature for stage I and stage II gastric cancer based on NET-related genes.

The prognostic marker elastin correlates with epithelial-mesenchymal transition and vimentin-positive fibroblasts in gastric cancer

Elastin (ELN) fibers are essential constituents of the tumor microenvironment of gastric cancer (GC). However, few studies have investigated the clinical prognostic significance of ELN in GC. We screened for molecular markers that were highly related to distant metastasis by transcriptome sequencing. The Cancer Genome Atlas (TCGA) and Harbin Medical University (HMU) validation cohorts were used to validate ELN expression and to explore molecular mechanisms. Immunohistochemistry for ELN, vimentin (VIM), and fibroblast activation protein, and elastic fiber-specific staining were used to evaluate the relationship between ELN and prognosis. R studio was used to construct a nomogram prognostic model. In this study, we found that ELN mRNA levels were significantly higher in cancer tissues and were associated with poor prognosis in TCGA and HMU patients. Gene set enrichment analysis showed that ELN was mainly enriched in the epithelial-mesenchymal transition (EMT) pathway. The mRNA expression of ELN was positively correlated with fibroblast molecular markers, especially VIM. For validation, we collected a tissue microarray containing 180 pairs of samples. We found that ELN was positively correlated with VIM expression in cancer tissue but not in paracancerous tissues by immunohistochemistry staining. Univariate and multivariate analyses showed that the expression of ELN and lymph node metastasis rate were independent predictors for overall survival. Moreover, a nomogram model was used to evaluate the risk of death by combining the expression of ELN and lymph node metastasis rate. ELN may play an important role in the progression of GC by regulating EMT and is a useful prognostic indicator in predicting the prognosis of GC.

Lymph Node Metastasis-Related Gene ITGA4 Promotes the Proliferation, Migration, and Invasion of Gastric Cancer Cells by Regulating Tumor Immune Microenvironment

The Integrin Subunit Alpha 4 (ITGA4) plays important roles in cancers pathogenesis. However, the expression and association with clinicopathological and survival probability have not been previously assessed in gastric cancer (GC). Protein expression of ITGA4 was assessed in TMA using immunohistochemistry and correlated with clinicopathological factors and survival. The mRNA expression of ITGA4 was also assessed in the HMU-GC cohort. Bioinformatics function analysis was conducted through GSEA. The “CIBERSORT” package was used for immune infiltration analysis. “SvyNom” package is used to construct prognosis model. ITGA4 knock down using shRNA. The evaluation of cell function was performed by CCK-8 and Transwell invasion and migration experiments. ITGA4 was significantly associated with N classification ( $P=0.031$ ), tumor location ( $P=0.033$ ), WHO classification ( $P=0.007$ ), and poor prognosis in mRNA level. GSEA analysis of the validation cohort suggested that ITGA4 was associated with macrophage infiltration. Immunohistochemistry showed that ITGA4 was associated with poor prognosis. Multivariate Cox regression analysis found that ITGA4 ( $P=0.045$ ) and lymph node metastasis rate ( $P=0.026$ ) were independent prognostic factors and could construct a prognosis model. ITGA4 knockdown cell line significantly reduced the ability of proliferation, invasion, and metastasis. ITGA4 is associated with patient survival in GC and may be an important prognostic biomarker.

Prognostic importance of the preoperative New-Naples prognostic score for patients with gastric cancer

BACKGROUND

The wide applicability of the Naples prognostic score (NPS) is still worthy of further study in gastric cancer (GC). This study aimed to construct a New-NPS based on the differences in immunity and nutrition in patients with upper and lower gastrointestinal tumors to help obtain an individualized prediction of prognosis.

METHODS

This study retrospectively analyzed patients who underwent radical gastrectomy from April 2014 to September 2016. The cutoff values of the preoperative neutrophil-to-lymphocyte ratio (NLR), lymphocyte-to-monocyte ratio (LMR), serum albumin (Alb), and total cholesterol (TC) were calculated by ROC curve analysis. ROC and t-ROC were used to evaluate the accuracy of the prognostic markers. The Kaplan-Meier method and log-rank test were used to analyze the overall survival probability. Univariate and multivariate analyses based on Cox risk regression were used to show the independent predictors. The nomogram was made by R studio. The predictive accuracy of nomogram was assessed using a calibration plot, concordance index (C-index), and decision curve.

RESULTS

A total of 737 patients were included in training cohort, 411 patients were included in validation cohort. ROC showed that the New-NPS was more suitable for predicting the prognosis of GC patients. NPS = 2 indicated a poor prognosis. Multivariate analysis showed that CEA (P = 0.026), Borrmann type (P = 0.001), pTNM (P < 0.001), New-NPS (P < 0.001), and nerve infiltration (P = 0.035) were independent risk factors for prognosis.

CONCLUSION

The New-NPS based on the cutoff values of NLR, LMR, Alb, and TC is not only suitable for predicting prognosis but can also be combined with clinicopathological characteristics to construct a nomogram model for GC patients.

FSIP1 Is Associated with Poor Prognosis and Can Be Used to Construct a Prognostic Model in Gastric Cancer

Gastric cancer (GC) is one of the most common upper gastrointestinal malignant tumors, and the incidence of the GC shows an increasing trend in the past years. Finding more sensitive markers will help to reveal the mechanism of GC progression and clinic diagnoses. This study first analyzed the mRNA expression level of FSIP1 in TCGA GC samples and the significance in predicting the prognosis. KEGG and GO analyses were used to explore the molecular mechanism of FSIP1 in GC progression. This study further retrospectively analyzed 166 clinical samples of GC from Harbin Medical University Cancer Hospital and evaluated the expression level of FSIP1 by immunohistochemistry. Kaplan-Meier and Cox multivariate analysis was used to investigate the prognostic value of FSIP1 expression in GC patients. We also identified correlations between FSIP1 and clinicopathological characteristics. This study found that the mRNA level of FSIP1 was significantly upregulated in GC compared with nontumor specimens and correlated with poor prognosis. Immunohistochemistry confirmed the results of bioinformatics analysis of the TCGA GC database. FSIP1 was associated with pTNM pathological stage, tumor location, and neural invasion. In addition, multivariate Cox regression analysis showed that FSIP1, T classification, and N classification were independent posterior factors of patients and could be combined with pathological features to construct a nomogram prognostic model. Overall, our results suggest that FSIP1 is expected to be an independent prognostic indicator of GC.