The order of surgery and chemotherapy matters: Multimodality therapy and stage-specific differences in survival in gastric cancer.

BACKGROUND AND OBJECTIVES: Multimodality treatment improves survival for gastric cancer (GC). However, the effect of treatment sequence by stage remains unclear. We aim to compare outcomes between patients receiving neoadjuvant(neoadj) and adjuvant chemotherapy (adj). METHODS: Nonmetastatic GC patients with clinical stage ≥ T2N0 who underwent both resection and neoadj or adj were identified using the National Cancer Database (2005-2014). Multivariable Cox regression analyses were performed on propensity score-matched (PSM) cohorts stratified by stage to compare overall survival (OS). RESULTS: We identified 11 984 patients; 55% stage I (SI), 76% stage II (SII) and 57% stage III (SIII) received neoadj. Unadjusted analysis showed worse survival among SI neoadj patients (hazard ratio [HR] 1.195, confidence interval [CI] 1.04-1.38) and improved survival for SII (HR 0.93 CI 0.87-0.998) and SIII (HR 0.75, CI 0.68-0.84). After PSM, SI patients with neoadj had worse OS with increased risk of death compared to Adj (HR 1.186, CI 1.004-1.402). SII patients had no difference in OS (HR 0.98, CI 0.91-1.07) and SIII patients had improved OS (HR 0.78, CI 0.69-0.90). CONCLUSIONS: In patients who received surgery and chemotherapy, the benefit of neoadj was limited to SIII with worse survival for SI. A clinical trial to examine the optimal sequence of chemotherapy is warranted.