354. THE PARADIGM OF INACCURACIES IN STAGING EARLY OESOPHAGEAL ADENOCARCINOMA

Abstract

Abstract The clinical value of multiple staging investigations for high-grade dysplasia (HGD) or early adenocarcinoma of the oesophagus (EOC) is unclear. A single centre prospective cohort of patients treated for EOC between 2000 and 2019 was analysed. This coincided with a transition period from oesophagectomy to endoscopic mucosal resection (EMR) as the treatment of choice for early cancers. All patients were staged with computed tomography (CT), endoscopic ultrasound (EUS), and 2-deoxy-2-[18F]fluoro-d-glucose (FDG) positron emission tomography (PET)-CT. The aim of this study was to assess the accuracy of these investigations and impact on clinical management. 297 patients with HGD or EOC were included (EMRn = 184; oesophagectomy n = 113 [preceded by EMRn = 23]). Staging patterns for both groups were similar throughout staging modalities (CT, EUS, PET-CT), showing low overall accuracy. In the EMR group, over-staging that might have changed management occurred in 10 patients (3.6%)for T-staging (T2) and 14 patients (5.0%)for positive lymph node-staging (LNM). In the oesophagectomy alone group LNM identified following surgery was rare (8.9%) and staging sensitivity for these surgically treated patients was low (CT 12.5%, EUS 12.5%, PET-CT 0.0%). Overall, PET-CT and EUS changed decision-making in less than 1%of patients with a clear CT and low risk histology criteria on EMR specimen. Staging accuracy is low in EOC and the addition of EUS and PET-CT does not change patient decision making significantly. Histology obtained from EMR specimens contain important predicting factors that do change patient management, proving to be an excellent staging tool rather than imaging techniques, hence, EUS and PET-CT should only be used selectively in patients being assessed for EOC.