Which Lymph Node Staging System Better Predicts Prognosis in Patients With Gastric Carcinoma? A Comparative Study Between 3 Different Lymph Node Classifications for Resected Gastric Cancer in a Western Tertiary Center

Introduction: Gastric cancer (GC) is an aggressive disease with high mortality rates. Lymph node (LN) staging of GC is a major source of controversy. The aim of this study is to compare the prognostic value of 3 different LN classifications for patients with resected GC: the eighth TNM staging system, lymph node ratio (LNR, ratio between positive and total LN) and a new anatomic-based classification (Choi classification). Materials and Methods: A retrospective study of all cases of GC resected in a tertiary hospital in Spain (n=377). Clinical data were collected; histologic slides were reviewed; and univariate and multivariate analyses of disease-free survival (DFS) and overall survival (OS) were performed. Results: In all, 315 patients fulfilled inclusion criteria. Univariate analysis showed that all classifications were significantly associated with tumor death and progression (P<0.001). All staging systems were independent prognostic factors for DFS. Area under the curve ratios for Choi, N stage, and LNR classifications were 0.738, 0.730, and 0.735, respectively. TNM and LNR classifications were independent prognosticators for OS, while Choi classification was an independent factor only in patients with >= 16 LN resected. Area under the curve ratios for Choi, N stage, and LNR classifications were 0.707, 0.728, and 0.732, respectively. Kaplan-Meier curves depending on LNR classification showed the best patient stratification for both OS and DFS. Conclusions: The 3-staging systems had similar prognostic performance, but LNR-based classification stratified patients better. Further studies are needed to evaluate the impact of the number of LN examined, cutoff values, and anatomic extent of LN disease in GC.

Automated summary

This study compared the prognostic value of 3 different LN classifications for patients with resected GC, showing that LNR-based classification stratified patients better for prognosis assessment. Further studies are needed to evaluate the impact of the number of LN examined, cutoff values, and anatomic extent of LN disease in GC.