Background: On the basis of SEER data, in which most gastric cancer patients had limited lymph node dissection, node ratio-based staging system (TNrM) has been shown to have better accuracy than the AJCC TNM system. This study is to validate the result with patients from Taiwan, where D2 lymphadenectomy is routinely performed.
Patient and method: A total of 1405 consecutive gastric cancer patients treated at National Taiwan University Hospital between 1998 and 2010 were included. To evaluate the performance of the AJCC system, each TNM stage was stratified by TNrM stages. The homogeneity of patients' survival across TNrM strata was evaluated using the log-rank test. The performance of the TNrM system was evaluated with the same approach.
Results: Five of the 7 evaluable AJCC stages (IA, IIA, IIIA, IIIB, and IIIC) contained TNrM subgroups with statistically heterogenous survival (P=0.003, 0.04, 0.002, 0.04, and <0.001, respectively). Thirty-six percent of patients (506/1405) were misclassified by the AJCC TNM system. However, of the assessable 6 TNrM stages, none of the AJCC subgroups showed significantly heterogenous survivals (P>0.05). About 19% of patients (264/1405) were misclassified by using the TNrM system.
Conclusions: Lymph node ratio significantly decreases the stage migration caused by inadequate examined lymph nodes. The advantage of TNrM was validated with a patient cohort from the Eastern medical center.