ABSTRACT
BACKGROUND: Our aim was to evaluate the prognostic value of the number of lymph nodes examined (eLNs) in patients with node-negative gastric cancer (GC) and further to adjust the American Joint Committee on Cancer (AJCC) 8th staging system based on the number of eLNs. METHODS: Node-negative GC patients diagnosed during 1988-2015 from the Surveillance, Epidemiology, and End Results (SEER) database were included. On the basis of a primary cohort of 4159 node-negative GC patients, we built the adjusted AJCC 8th staging system, which was then internally validated by a bootstrap method, and externally validated with an independent cohort of 5565 node-negative GC patients. RESULTS: The median number of eLNs was 10. For the training set, the optimal eLNs thresholds were determined to be 9 for node-negative GC patients. The adjusted AJCC 8th staging system for node-negative GC patients based on the number of eLNs had a significantly higher Harrell's concordance index than the initial AJCC 8th staging system (C-index, 0.635 versus 0.616; P < 0.001). Thus, the adjusted AJCC 8th staging system had superior prognostic stratification. Similar results were found in the validation set. CONCLUSIONS: For node-negative GC patients in the United States, the adjusted AJCC 8th staging system based on the number of eLNs predicted survival more accurately and discriminatively.