Introduction In Japan endoscopic submucosal dissection (ESD) is accepted as a safe and effective treatment for early oesophageal cancer. Experience in the UK remains limited and oesophagectomy is still the gold standard. The aim of this prospective single centre pilot study was to evaluate the safety and clinical outcomes of oesophageal ESD in a UK setting.
Methods Between July 2008 and November 2011 the regional upper GI MDT for North Wales and Cheshire considered 14 patients with early oesophageal cancer (T1N0M0) (n=11) and high grade dysplasia (n=3) for ESD after full staging. All patients underwent trimodal endoscopy (autofluorescence, narrow band imaging, magnification, and chromoendoscopy) to assess the lesion and depth of invasion. Informed consent was obtained after full discussion and counselling as to alternative treatment options. Standard ESD technique was used, whereby the lesion was isolated by circumferential cutting using a flush and IT2 knife after marking the edges and raising with submucosal injection; followed by dissection. Specimens were staged according to the Kikuchi classification. Patients with residual Barrett's (n=5) had radio-frequency ablation after ESD to reduce the risk of metachronous cancer. Data were collected prospectively and audited by an independent group.
Results Of the 14 cases (nine male, five female; mean age 73 years), two were excluded as trimodal endoscopy showed evidence of deep submucosal infiltration and one patient declined treatment. Mean specimen size was 16 mm. Procedure time ranged from 120 to 210 min. Enbloc resection rate was 91%. R0 resection rate of the lateral and deep margins were 82% and 64% respectively (Abstract OC-146 table 1). There were no major complications, although one procedure was abandoned as the endoscopic field of view was obscured by bleeding. Mean hospital stay was 72 h. Procedure and disease specific mortality was zero. Over a median follow-up period of 20.5 months there was one recurrence. This occurred in a patient with incomplete resection of both lateral and deep margins at ESD. Those with R1 resection of the deep margins showed no evidence of recurrence.
Conclusion ESD is a safe and effective treatment with high cure rate for early oesophageal neoplasm, even when the endoscopist is in the steep part of the learning curve. ESD has the advantage of high enbloc resection rates and low risk of recurrence. In our opinion all patients in the UK with early oesophageal cancer and high grade dysplasia should have access to ESD as a standard treatment option.
Competing interests None declared.