Compartment-specific lymph node ratio improves prognostic stratification beyond conventional nodal staging in gastric cancer

Le résumé fourni par la source

Gastric cancer (GC) pN staging is based only on metastatic lymph node number, ignoring anatomical distribution. Lymphatic dissemination is hierarchical, leading to survival heterogeneity within the same pN category. We evaluate whether integrating compartment-specific nodal involvement improves prognostic discrimination. This retrospective study included 1,558 patients who underwent R0 gastrectomy with D2 lymphadenectomy between 2015 and 2018. Patients were randomly assigned to a training cohort ( n = 1067) and validation cohort ( n = 491). Compartment-specific lymph node metastasis rates (csLNR) in D1 and D2 compartments were combined to construct a csLNR staging system. A nomogram incorporating csLNR was developed and compared with conventional lymph node ratio (LNR)-based and pTNM(AJCC8th) staging systems. Five-year overall survival(OS) differed significantly across csLNR categories (89.1%, 66.3%, 43.5%, and 8.7% for csLNR0, csLNR1, csLNR2/3, and csLNR4, respectively; P < 0.001). Within pN2 and pN3 subgroups, csLNR further stratified patients into distinct prognostic groups with significantly different survival outcomes(pN2: 84.62%, 71.12% and 37.18% for csLNR0, csLNR1 and csLNR2/3; pN3: 62.10%, 46.82% and 13.33% for csLNR1, csLNR2/3 and csLNR4). csLNR remained independently associated with OS (HR = 10.47, 95CI%;6.77–16.20, P < 0.001). The csLNR-based nomogram showed improved discrimination compared with LNR-based and pTNM models in both training (C-index 0.804 vs. 0.778 vs. 0.759) and validation cohorts (0.768 vs. 0.744 vs. 0.739). Incorporating anatomical nodal metastatic distribution into quantitative nodal assessment improves prognostic stratification over pN staging. The csLNR model reveals heterogeneity within conventional nodal categories and facilitates individualized postoperative risk stratification in GC.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.