Endoscopic Ultrasonography (EUS) and Endoscopic Mucosal Resection (EMR) for Staging and Treatment of High-Grade Dysplasia (HGD) and Early Adenocarcinoma (EAC) in Barrett's Esophagus (BE)

Abstract

Aim: Endoscopic therapy can be curative for BE and HGD (Tis) or EAC confined to the mucosa (T1m). The risk of lymph node metastasis increases with the depth of mucosal invasion, and rises rapidly with submucosal invasion (T1sm). EUS has provided the highest accuracy for clinical staging of depth of invasion, but remains imperfect. We tested a staging strategy of an initial EUS, followed by EMR in patients with ≤T1m on EUS. Methods: We staged patients with HGD or EAC with high-frequency EUS (20 MHz). Patients with EUS stage of ≥T1sm were referred to surgery. Patients with EUS stage ≤T1m had cap-assisted EMR of focal lesions in long-segment BE or of all short segment BE. EMR specimens were evaluated by histology: m1 = tumor above the basement membrane (HGD); m2 = tumor invading the lamina propria; m3 = tumor invading the muscularis mucosae; sm = tumor invading the submucosa. Results: A total of 50 patients (40 M, 10 F), mean age 70 years (range 36-87 years) with HGD and possible EAC in BE were evaluated with EUS. In 8 patients, EUS indicated ≥T1sm. EUS was accurate compared to surgical pathology in 7/8 cases, but over-staged 1 patient with T1m disease. EMR was carried out in the remaining 42 patients (20 focal lesions, 22 short segment BE). EMR staging results compared with initial endoscopic biopsy diagnosis are shown in the table below. EMR was carried out as an outpatient procedure without serious complication. Subsequent treatment was based on EMR results and clinical status. In the 22 patients with m1 (HGD) on EMR, 1 had esophagectomy, and 21 had endoscopic follow-up, including 6 who received additional endoscopic therapy (EMR, photodynamic therapy, or argon plasma coagulation). None of the 21 developed metastatic cancer over a mean of 15.4 ± 9.0 months. In the 20 patients with EAC on EMR, 6 had esophagectomy, including 5/6 with sm invasion. In 12/14 patients, additional endoscopic therapy was carried out, and 1 patient received combined radiation and chemotherapy. None of the 14 developed metastatic cancer over a mean of 19.7 ± 8.9 months. Conclusions: Staging patients with HGD and EAC in BE can be performed accurately with EUS followed by EMR. Invasion of the deep mucosa and the submucosa can be definitively established. In medium-term follow-up, endoscopic therapy has been successful in patients with HGD and cancer limited to the mucosa.