Correlation of immune-inflammatory and tumor markers with lymph node metastasis in gastric cancer

To evaluate the combined predictive value of immune-inflammatory and tumor markers for lymph node metastasis (LNM) in gastric cancer (GC) patients.

We conducted a retrospective study of 207 GC patients who underwent radical gastrectomy. Based on postoperative histology, patients were categorized into LNM and non-LNM groups. Preoperative serologic levels of markers including carcinoembryonic antigen (CEA), carbohydrate antigen 19-9 (CA199), carbohydrate antigen 72-4 (CA724), Neutrophil-to-Lymphocyte Ratio (NLR), Platelet-to-Lymphocyte Ratio (PLR), Lymphocyte-to-Monocyte Ratio (LMR), Interleukin-6 (IL-6), and C-Reactive Protein (CRP) were collected. A nomogram prediction model was developed using multivariate logistic regression. Internal validation was performed using Bootstrap resampling, and external validation was conducted on an independent cohort of 97 patients.

LNM was present in 55 (26.6%) patients in the training cohort. Multivariate analysis identified preoperative levels of CEA (odds ratio [OR]=1.52, P<0.001), CA724 (OR=1.24, P<0.001), NLR (OR=2.86, P<0.001), and IL-6 (OR=1.97, P<0.001) as independent risk factors for LNM. The nomogram model incorporating these four factors demonstrated excellent discrimination, with an area under the curve (AUC) of 0.93. The model significantly outperformed conventional clinicopathologic indicators (P<0.001). Good calibration and clinical utility were confirmed by calibration curves and decision curve analysis, respectively. The model maintained strong predictive performance in both internal (AUC=0.92) and external (AUC=0.91) validation cohorts.

The combination of CEA, CA724, NLR, and IL-6 serves as an effective preoperative predictor of LNM in GC. The nomogram model based on these markers provides a reliable, non-invasive tool for individualized risk assessment and treatment planning.