Abstract
BACKGROUND: Patients undergoing radical gastrectomy demonstrate considerable variability in their prognoses, underscoring the urgent need for reliable biomarkers to inform personalized therapeutic strategies. This study seeks to develop and validate a novel prognostic model by combining preoperative immune, inflammatory, and nutritional biomarkers with computed tomography (CT) imaging features, thereby facilitating the prediction of outcomes in patients with gastric cancer who have undergone radical resection and aiding in the formulation of personalized clinical treatment strategies. METHODS: This retrospective study analyzed consecutive patients with a preoperative diagnosis of gastric cancer who underwent radical gastrectomy at two participating centers between January 2015 and December 2016. Based on predefined inclusion and exclusion criteria, eligible patients were randomly allocated to either a training cohort, which was used for model development and internal estimation of parameters, or a validation cohort, which served for independent testing of the model’s predictive performance. We assessed a range of preoperative hematological parameters and CT imaging features. Factors associated with overall survival (OS) were identified using least absolute shrinkage and selection operator (LASSO) regression analysis, and a prognostic model was subsequently constructed. RESULTS: A total of 393 patients were enrolled in the study and randomly allocated to the training and validation cohorts in a 7:3 ratio. The final prognostic model incorporated eight hematological indicators: white blood cell (WBC) count, hemoglobin (HB), total protein (TP), creatinine (Cr), neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), prognostic nutritional index (PNI), and systemic immune-inflammatory index (SII), in addition to CT staging characteristics. Time-dependent receiver operating characteristic (ROC) analysis of the risk scores yielded areas under the curve (AUCs) of 0.715, 0.740, and 0.736 for the training set, and 0.597, 0.631, and 0.657 for the validation set at 1, 3, and 5 years, respectively. High-risk patients had a substantially worse overall survival rate than low-risk patients, according to Kaplan-Meier analysis. CONCLUSION: The immune, inflammatory, and nutritional-CT (IIN-CT) model, which integrates preoperative immune, inflammatory, and nutritional biomarkers with CT imaging features, significantly improves the accuracy of preoperative prognostic predictions in gastric cancer patients. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12957-025-04155-9.