Abstract
BACKGROUND: To determine the optimal multimodal treatment strategy between perioperative chemotherapy (PEC), postoperative chemoradiation therapy (POCR), and postoperative chemotherapy (POC) in resected gastric cancer (GC) patients based on nodal status. METHODS: In this retrospective analysis, the National Cancer Database was used to identify resected non-metastatic GC (2006-2016). Patients were stratified by clinical nodal status-negative (cLN-) and positive (cLN+). In patients with cLN- disease who underwent upfront resection and were upstaged to pathological LN+, overall survival (OS) was compared between POC and POCR. In patients with cLN- and cLN+ disease, OS was compared between PEC, POCR, and POC. Kaplan-Meier survival estimate, log-rank test, and multivariable Cox proportional hazards analysis were performed. RESULTS: We identified 7827 patients (cLN- 4828; cLN+ 2999). On multivariable analysis in patients with cLN- disease who underwent upfront resection (n = 4314) and were upstaged to pLN+ disease (70%), POCR (n = 2300, aHR 0.78, 95% CI 0.70-0.87, p < 0.001) was associated with improved OS compared to POC (n = 907). No significant difference was noted between POCR (n = 766, aHR 1.11, 95% CI 0.88-1.40, p = 0.39) and POC (n = 341) in patients with pLN- disease. On multivariable analysis in all patients with cLN- disease, POCR (n = 3066) was significantly associated with improved OS (aHR 0.84, 95% CI 0.75-0.92, p < 0.01) compared to POC (n = 1248). No significant difference was noted between POCR (aHR 1.0, 95% CI 0.70-1.01, p = 0.958) and PEC (n = 514). These results remained consistent in patients with cLN+ disease (POCR = 1602, POC = 720, PEC = 677). CONCLUSION: Postoperative chemoradiation is associated with improved survival in GC patients upstaged from clinically node-negative disease to pathologically node-positive disease. Negative clinical nodal disease status is not a reliable indicator of pathological nodal disease.