Validation of staging systems for gastric cancer

Background. The two major staging systems for gastric cancer, the Japanese classification of gastric cancer (JCGC) and the International Union Against Cancer (UICC) TNM system, are periodically revised as a consequence of critical validation studies in light of newly accumulated clinical data. This study aimed to validate and improve upon the current versions for a better prognostic stratification of gastric cancer. Methods. One thousand and ten gastric cancer patients who underwent tumor resection were enrolled at the Kitasato University Hospital for staging validation. According to the JCGC stage, the patients consisted of stage IA (n = 453), IB (n = 185), II (n = 119), IIIA (n = 75), IIIB (n = 51), and IV (n = 127). Results. Regarding consistency between the JCGC and the UICC system, the results were: for patients in stage IA (100%), IB (98%), II (84%), IIIA (51%), IIIB (24%), and IV (64%). The JCGC system was superior to the UICC system for the prognostic stratification of stage IIIA, IIIB, and IV cancers; we therefore used the JCGC system for prognostic validation according to depth of invasion in cancers of the same stage. Stage II and IIIA cancers were heterogeneous for prognosis according to depth of invasion, and the outstanding difference was found between the muscularis propria (MP) and subserosa (SS), which are both classified as pT2 in the JCGC system. MP cancer represented an earlier property of gastric cancer rather than an advanced one. A proposed novel staging system adjusted for this heterogeneity provided a clearer stratification of prognosis with a homogeneous prognostic distribution within each stage. Conclusion. Our findings revealed that invasion into the MP has an earlier propensity than expected, and a novel staging system taking this into account may provide a better stratification of prognosis than the current systems.