Survival analysis comparing neoadjuvant chemoradiotherapy versus perioperative chemotherapy for locoregional gastric cancer.

Abstract

e16097 Background: Surgical resection is established as the only potentially curative approach for patients with early stage and regional gastric cancer. With a high risk of recurrence that ranges between 40% to 60% for stage II and III disease, combination therapies including chemotherapy and radiation that complement surgery were incorporated as standard of care for Stage >IB disease with clear survival benefit over surgery alone. While the National Comprehensive Cancer Network (NCCN) guidelines recommend neoadjuvant chemoradiation (NCR) or perioperative chemotherapy (PC) for clinical Stage T2 or higher, there has been no consensus on the optimal treatment approach for these patients. Our study sought to compare survival differences between the two therapies. Additionally, we analyzed predictive factors and shifting trends in the choice of treatment over the period of our study i.e., 2010 to 2017. Methods: The National Cancer Database was used to analyze patients with non-metastatic regional gastric cancer diagnosed from 2010 to 2017 staged as clinical T2 or higher. Demographic data and clinicopathological tumor characteristics were assessed using chi-square analysis to assess baseline differences between both groups. The primary outcome was to determine differences in overall survival (OS) using Kaplan Meier (KM) and Cox proportional hazards analysis. The secondary outcome was to evaluate patient related factors associated with the choice of treatment. Results: N = 7475 patients with non-metastatic clinical T2+ gastric cancer were included in the study (NCR: n = 5394, 72.16% vs. PC: n = 2081, 27.84%). The majority of patients were of male sex (75.9%) and within the 50-69 age group (64.5%) for both groups. T3 clinical staging was the most prevalent among both groups (78.0% and 70.3% respectively). Patients receiving PC had longer median OS (49.480 months) than those undergoing NCR (40.970 months). KM curve showed a higher cumulative survival with PC over NCR at 1-year, 3-year, and 5-year endpoints (0.93 vs. 0.85, 0.58 vs. 0.54, and 0.46 vs. 0.40, p<0.0001). Moreover, on multivariate cox regression analysis, OS was inferior in the NCR group (HR 1.57, p<0.001) compared to PC after adjusting for significant covariates. In a subgroup analysis for clinical T3 and T4 tumors, NCR remained associated with worse OS (HR 1.83, p <0.0001) compared to PC. Lastly, patients in the 70-79 and 80-100 age group preferred NCR over the PC approach (OR 0.737 and OR 0.50, p 0.035 and p 0.008, respectively) when compared to the younger population. Cochran-Armitage Trend Test revealed NCR (p <0.0001) and PC (p <0.0001) had an increased trend by year. Conclusions: In our study, PC showed clear survival benefits over NCR in patients with locoregional gastric cancer. Future large-scale clinical trials would aid in adding to our findings and setting guidelines for optimal management of these patients.