Background: Accurate staging is essential in cancer care. The American Joint Committee on Cancer (AJCC) TNM staging system is commonly used but is subject to the risk of stage migration. Recent literature suggests that the log odds of positive lymph nodes (LODDS) and positive lymph node ratio (LNR) may have superior predictive values and are considered alternatives to the N-category. However, their predictive performance in gastric signet ring cell carcinoma (GSRC) remains vague. This study aims to explore the association between three lymph node (LN) staging systems (AJCC N-category, LODDS, and LNR) and outcomes in GSRC, and assess the predictive power.
Methods: Eligible patients with GSRC from 2004 to 2015 were collected from the Surveillance, Epidemiology, and End Results database. The time-dependent receiver-operating characteristic (ROC) analysis, area under the curve (AUC), and integrated discrimination improvement (IDI) were used to assess the predictive performance of the three LN stages (AJCC N-category, LODDS, and LNR).
Results: In the multivariate analysis of all GSRC patients and the subgroup of patients with ≤ 15 LNs examined, both the LODDS and LNR were significant survival prognostic factors. The time-dependent ROC curves of the LODDS and LNR exhibited higher sensitivity and specificity when compared to the N-category curve. The AUCs at 1, 3, and 5 years demonstrated that the predictive performance of LODDS and LNR was significantly better than the N-category (all P < 0.05). IDI of LODDS and LNR also showed sufficient fit and attractive net benefit in prediction and clinical application.
Conclusions: LODDS and LNR were remarkable prognosticators for survival in GSRC patients. Their predictive performance was better than that of the N-category, indicating that LODDS and LNR could ameliorate the predictive precision of survival risk and could replace the N-category in predicting the outcomes of GSRC patients.
Keywords: Gastric signet ring cell carcinoma; Log odds of positive lymph nodes; Lymph node ratio; Prognosis.
© 2025. The Author(s).