Predictors of Positive Margins After Definitive Resection for Gastric Adenocarcinoma and Impact of Adjuvant Therapies

Abstract

Positivemargins after definitive resection in gastric adenocarcinoma are associated with inferior outcomes. There are few randomized data to guide optimal adjuvant therapy after positive margins. Using the National Cancer Database, we identified 24,619 nonmetastatic gastric adenocarcinoma patients who received diagnoses from 2004 to 2013 and underwent definitive resection to analyze for predictors of positive surgical margins. Of these patients, 2754 had positive margins (11.2%). Multivariate prevalence ratios were used to determine predictors. Survival analyses were performed with a Cox proportional hazards model by use of several methods of propensity score analysis. Increasing T and/or N category, high grade, and lymphovascular invasion predicted higher rates of positive margins. Asian race, treatment at an academic center, and robotic surgery predicted lower rates of positive margins. Among positive-margin patients with adjuvant treatment (n=1021), with a median follow-up period of 55months, age, comorbidity score, nodal disease, and T4 disease predicted for worse overall survival (OS). Treatment at an academic center was associated with improved OS. Use of adjuvant concurrent chemoradiation therapy (CCRT) was associated with higher OS compared with chemotherapy alone after positive margins (hazard ratio, 0.72; 95% confidence interval, 0.58-0.91; P=.005) after propensity matching adjusting for predictors of OS. The 2-year and 3-year OS for positive-margin patients with chemotherapy alone was 43% and 29%, respectively, compared with 53% and 38%, respectively, with adjuvant CCRT. The log-rank P value for survival was .0015. Stage, race, treatment center, and surgery approach predict for margin status after resection. Adjuvant CCRT after positive margins is associated with improved OS after accounting for available clinical variables.