RESUMO
PURPOSE: Current gastric cancer staging systems overlook the anatomic extent of metastatic lymph nodes (AEMLNs). This study aimed to analyze the prognostic impact of AEMLNs on gastric cancer (GC). METHODS: GC patients with metastatic lymph nodes (MLNs) undergoing curative surgery were retrospectively reviewed and assigned to perigastric (MLNs in station 1-6, PG) and extraperigastric group (7-12, with or without MLNs in PG area, EPG). Overall survival (OS), disease-free survival (DFS) and recurrence patterns were compared before and after 1:1 propensity score matching (PSM). RESULTS: 662 patients were enrolled, 341 (51.5%) and 321 (48.5%) of whom were in the PG and EPG, respectively. After PSM (n = 195), EPG showed poorer 5-year OS (43.4% vs 54.5%, p = 0.014) and DFS (65.0% vs 73.4%, p = 0.068) than PG. EPG had higher incidence of peritoneal recurrence (PR) than PG (19.4% vs 7.4%, p = 0.002). Multivariate analysis identified AEMLNs as prognostic factor for OS [HR = 1.409, 95% confidence interval (CI) 1.062-1.868), DFS (HR = 1.600, 95% CI 1.059-2.416) and PR (HR = 3.708, 95% CI 1.685-8.160). CONCLUSIONS: The anatomic extent of metastatic lymph nodes has an independent prognostic role for GC. Including this element may improve the accuracy of current staging systems.