Construction and validation of a prognostic model for gastric cancer patients with tumor deposits

Background . Tumor deposits (TD) was a significant risk factor impacting the prognosis of patients diagnosed with gastric cancer (GC), yet it was not currently incorporated into TNM staging systems. The objective of this research was to develop a predictive model for assessing the prognosis of patients with TD-positive GC. Methods . Retrospective analysis was performed on the data of 4,972 patients treated for GC with D2 radical gastrectomy at Wannan Medical College's Yijishan Hospital between January 2012 and December 2021. The patients were categorized based on the number of TD (L1: 1, L2: 2-3, L3: >= 4) and the anatomical location of TD (Q1: single area, Q2: multiple areas). In a 3:1 ratio, patients were randomly assigned to one of two groups: training or validation. Results . The study included a total of 575 patients who were divided into the training group ( n = 432) and validation group ( n = 143). Survival analysis showed that the number and anatomical location of TD had a significant impact on the prognosis of patients with TD-positive GC. Univariate analysis of the training group data revealed that tumor size, T-stage, N-stage, histological grade, number and distribution of TD, neural invasion, and postoperative chemotherapy were associated with prognosis. Multivariate Cox regression analysis identified poor histological grade, T4 stage, N3 stage, number of TD, neural invasion, and postoperative chemotherapy as independent prognostic factors for GC patients with TD. A nomogram was developed using these variables, demonstrating well predictive ability for 1, 3, and 5-year overall survival (OS) in the validation set. The DCA curve shows that the constructed model shows a large positive net gain compared to the eighth edition Tumour, Node, Metastasis (TNM) staging system. Conclusion . The prognostic model developed for patients with TD-positive GC has a higher clinical utility compared to the eighth edition of TNM staging.