When we evaluated the homogeneity of groups for the seventh edition of the TNM staging system, N3a (7-15 metastatic lymph nodes) and N3b (>15 metastatic lymph nodes) were considered as 1 group because the seventh edition already categorized them as 1 group (N3). Two stages (stages IIB and IIIA) had homogeneity in the seventh edition, whereas only 1 stage (stage II) demonstrated homogeneity in the sixth edition of the TNM staging system.1 Therefore, we concluded that the seventh edition of the TNM staging system provides more detailed grouping. However, as in our discussion, the data suggest that N3 (>6 metastatic lymph nodes) in the seventh edition is too broad a classification. If we divide the N3 group into N3a and N3b, patients with T1, T3, and T4a disease demonstrated significantly different survival rates (77.2% vs 35.0% [P = .0161], 42.2% vs 24.3% [P <.001], and 32.4% vs 18.4% [P <.001], respectively), although patients with T2 and T4b disease did not show significant survival differences (59.2% vs 39.9% [P = .063] and 6.1% vs 17.2% [P = .978], respectively) as shown in Figure 5 (bottom) in our article.1 We need a larger database to answer this issue. Therefore, the American Joint Committee on Cancer/International Union Against Cancer TNM committee for gastric cancer decided to encourage the use of the N3and N3b classifications for data collection and it is expected that data from high-volume centers throughout the world will help future revisions of the TNM classification.
Both the sixth and seventh editions of the TNM staging system use numeric classification of the N classification and do not define the extent of surgical dissection. Only the minimal number of lymph nodes required for dissection is described (>15 metastatic lymph nodes). Therefore, in this stage migration analysis, the extent of surgery is not an issue, although in our institute, for patients with clinical T1 cancers, the lymph nodes around the stomach, left gastric artery, common hepatic artery, and celiac trunk were dissected and D2 lymph node dissection was performed for other T classifications.
Lee et al reported that as long as at least 15 lymph nodes were examined, the number of lymph nodes examined did not significantly affect lymph node staging of gastric cancer.2 We reanalyzed our data in patients who had at least 16 lymph nodes examined, and the resultant survival rates in each stage were not found to be significantly different from those reported in our report,1 and the statistical differences were maintained.