467. DISTRIBUTION OF LYMPH NODE INVOLVEMENT IN PATIENTS WITH GASTRO-ESOPHAGEAL JUNCTION ADENOCARCINOMA

Abstract

Abstract Background Curative cancer surgery relies on understanding the lymphatic spread of the tumour to guide lymphadenectomy. The lymphatic involvement in patients with esophageal squamous cell carcinoma (SCC) has been evaluated extensively in large cohort studies from the East where SCC predominates, but the pattern of lymph node spread in gastro-esophageal junction (GEJ) adenocarcinoma is less clear. The primary aim of this study was to compare differences in the pattern of lymph node involvement according to the Siewert classification. Methods Consecutive patients who underwent curative resection for GEJ adenocarcinoma between February 2010 and April 2023 were analysed from a prospectively maintained database. Patient demographic data, neo-adjuvant treatment, operation type and histopathological report data were collected and analysed according to the pathological Siewert classification of GEJ cancer. Results 807 patients who underwent curative surgery for GEJ adenocarcinoma were analysed. The median lymph node yield was 26 (5-81). Comparing positive para-esophageal node yields, Type I and II tumours were similar but both significantly higher than Type III. For positive para-gastric nodes, Type 2 and 3 tumours were similar but both significantly higher than type 1. The significance of these comparisons is presented in table 1. There were no differences comparing other specific nodal groups. Evaluating the effect of node involvement on overall survival positive node status and positive hepatic artery nodes were found to be independent factors adversely affecting survival. Conclusion Siewert type I tumours have greater proximal lymph node involvement compared to type III tumours with higher proportions of positive para-esophageal nodes. Type III tumours on the other hand have higher proportions of para-gastric nodes. Type II tumours have similar para-esophageal node involvement to type 1 tumours and similar para-gastric involvement to type III tumours. Involvement of more specific nodal groups does not seem to vary between the sub types.