Stage Migration in Gastric Cancer: Its Influence on Survival Rates

TNM staging is commonly used in gastric cancer patients. However, the diagnostic information on positive lymph nodes available for this staging depends on the dissection performed. For patients who have undergone D2 dissection, information on the N2 tier is available; this is not the case after a D1 dissection. This implies that TNM stages derived for D2 dissected patients are systematically different from those for D1 patients. Consequently the comparison of stage-specific survival rates between D2 and D1 patients is confounded by this systematic staging difference. The purpose of this paper was to study a cohort of D2 dissected patients with meticulous documentation of the lymph node status to determine the effect of neglecting the N2 information; for this purpose they were also staged as D1 dissected patients. How large is the stage migration when D2-TNM staging is compared to D1-TNM staging, and what are the consequences of this migration on stage-specific survival rates? For 291 D2 patients within the Dutch D1-D2 randomized trial, T and N classifications were available as well as 5-year survival data. Therefore the DP-TNM stages were known and the 'simulated' D1-TNM stages could be derived by neglecting the lymph node information of the N2 tier. Consequently, stage migration and its impact on stage-specific survival could be studied. Stage migration occurred in 86 cases (30%). For stages II and IIIa in particular the observed stage-specific 5-year survival rates for the D2-TNM stages were higher than those for the simulated D1-TNM stages: there was a difference of 7% for stage II and 13% for stage IIIa patients. Comparison of stage-specific survival rates between Japan and Western countries for gastric cancer patients is confounded by stage migration. This is due to the fact that N2 lymph node information is available for D2 patients but not for D1 patients.