BACKGROUND: The current and the previous editions of the tumor-node-metastasis (TNM) system for gastric cancer (GC; TNM8 and TNM7) have a high risk of stage-migration bias when the node count after gastrectomy is suboptimal. Hence, they are possibly not the optimal staging systems for GC patients. This study aims to compare the TNM with two systems less affected by the stage-migration bias, namely, the lymph nodes ratio (LNR) and the log odds of positive lymph nodes (LODDS), to assess which one is the best in stratifying the prognosis of GC patients.
METHODS: The sample study included 1221 GC patients. Two 7-cluster staging systems based on the combination of pT categories and LNR and LODDS categories (TLNR and TLODDS) were compared with the two last editions of TNM, using the Akaike information criteria, the Bayesian information criteria, and the receiver operating characteristic (ROC) curve graphs. Further validation on an independent sample of 251 patients was carried out.
RESULTS: The univariable and multivariable analyses and the ROC curves detected an advantage of the TLNR and TLODDS systems over the TNM. The TLNR and TLODDS showed the best accuracy both in the subgroup of patients with ≥16 nodes examined. The results were confirmed in the validation analysis.
CONCLUSIONS: TLNR and TLODDS staging systems should be considered a valid implementation of the TNM for the prognostic stratification of GC patients. If these results are confirmed in further studies, the future implementation of the TNM should consider the introduction of the LNR or the LODDS along with the number of metastatic nodes.
- Multivariate Analysis
- Neoplasm Staging/methods
- ROC Curve
- Retrospective Studies
- Stomach Neoplasms/diagnosis