RAMIE Versus MIE for Resectable Esophageal Cancer, a Randomized Controlled Trial (ROBOT-2 Trial).

Aim of the study This is a randomized controlled parallel-group, superiority trial comparing RAMIE to MIE with intrathoracic anastomosis (Ivor-Lewis) in patients with resectable esophageal adenocarcinoma or adenocarcinoma of the gastroesophageal junction in the Western World.

Methods Objectives Patients with resectable esophageal adenocarcinoma or adenocarcinoma of the gastroesophageal junction are randomized at the outpatient department to either (a) robot-assisted minimally invasive esophagectomy (RAMIE) or (b) conventional minimally invasive esophagectomy (MIE). The objective is to evaluate the extent of lymph node dissection, efficacy, risks, survival and cost-effectiveness of RAMIE as an alternative to MIE as treatment for esophageal adenocarcinoma or adenocarcinoma of the gastroesophageal junction. We hypothesize that RAMIE leads to an improved lymph node dissection compared to MIE.

Study design This is a multicenter investigator-initiated and investigator-driven randomized controlled parallel-group, superiority trial comparing RAMIE to MIE. This study is conducted in accordance with the principles of the Declaration of Helsinki and Good Clinical Practice Guidelines. The independent ethics committee of the University Medical Center of the Johannes Gutenberg University, Mainz, Germany has approved the study. Written informed consent will be obtained from all participating patients. All centers participating in the ROBOT-2 trial have extensive experience in minimally invasive esophageal surgery and have an experience with at least 50 MIE an 50 RAMIE procedures performed.

Clinical trial monitoring will be conducted by an independent data monitor from the Johannes Gutenberg University, Mainz, Germany. A Data safety monitoring board (DSMB) will evaluate safety for patients included within this trial. All outcomes will be evaluated by a (blinded) external independent data committee board for the Upper GI International Robotic association (UGIRA).

Study protocol Patients are informed about the trial by one of our surgeons at the outpatient department. After receiving the information, all patients get one week time to consider their consent. After obtaining informed consent, randomization, with concealment of allocation, is done centrally by an online randomization program. There is no blinding for the patient, surgeon and coordinating researcher because this is difficult in daily practice. However the UGIRA committee and DSMB are blinded to the allocated intervention. This study is funded by Intuitive surgical Inc., Sunnyvale, CA, USA. Neoadjuvant (radio)chemotherapy will be administered according to current international policy. Multiple esophageal cancer biopsies for pathological analysis will be obtained through esophagogastroscopy of which 4 biopsies will be snap frozen and stored for translational research. All resection specimens will be preserved and stored (biobank) for translational research. The study will start on April 1st 2020. Inclusion will take approximately 2 years. Follow up for each patient will be 5 years. Total duration of the study will be 7 years.

Surgery All procedures (RAMIE or MIE) will be carried out by experienced surgeons with experience of at least 50 MIE and 50 RAMIE procedures. All patients will receive an epidural catheter to provide adequate postoperative analgesia. Patients will be intubated with a left-sided double-lumen tube to enable selective desufflation of the right lung during the thoracic phase in both procedures.

Antibiotic prophylaxis (Ampicillin 2000 mg and Sulbactam 1000mg) will be administered 30 minutes prior to incision. A thoracic drain will be positioned in the right hemithorax at the end of the procedure. Extubating will take place in the operating theater directly postoperatively and hereafter all patients will be admitted to the intensive care unit (ICU) for hemodynamic and respiratory monitoring. Hemodynamical and respiratory stable patients were discharged towards the surgical ward. All patients were placed on a nil-by-mouth routine for the first 3 days postoperatively. In absence of clinical signs of anastomotic insufficiency, patients started with sips of water and the oral intake was gradually increased to solid food. There was no enhanced recovery program.

Surgical procedure: RAMIE The RAMIE technique using the 4 arm daVinci Xi system was described previously (daVinci Xi system, Intuitive Surgical Inc., Sunnyvale, CA, USA).20 For the abdominal phase, the patient is placed in supine position. Robotic trocars positions are shown in figure 2a. The lesser omentum is opened and transected closely to the liver, until the left crus of the diaphragm is reached. Hereafter, the greater gastric curvature is dissected. An abdominal lymphadenectomy is performed including lymph nodes surrounding the hepatoduodenal ligament, the celiac trunk, along the left gastric and splenic artery and the lesser omental lymph nodes. The left gastric artery and vein are ligated with robotic Hem-o-lok and transected at their origin. The gastric conduit is created at the level of the crow's foot using a (robotic) endostapler .

For the thoracic phase, the patient is positioned in the left lateral decubitus position, tilted 45° towards the prone position (semi-prone). Trocars positions are shown in figure 3a. The robotic system is brought into the field at the dorsocranial side of the patient. After incision and installation of the operation robot and selective desufflation of the right lung, the pulmonary ligament is divided. Hereafter, the parietal pleura is dissected at the anterior side of the esophagus from the diaphragm up to the azygos arch. The azygos vein is ligated with robotic Hem-o-lok and divided. Dissection of the parietal pleura is continued above the azygos arch to establish dissection of the right paratracheal lymph nodes. At the posterior side of the esophagus, the parietal pleura is dissected cranially to caudally along the azygos vein, including the thoracic duct. The thoracic duct is clipped with robotic Hem-o-lok to prevent chylous leakage. The esophagus is resected en bloc with the surrounding mediastinal lymph nodes. The gastric conduit is pulled up and the specimen is removed through a small incision (mini-thoracotomy) at the location of the trocar in the 6th intercostal space. Continuity is created with a circular stapled esophago-gastrostomy, which is routinely oversewn with a V-Lock (Medtronic). An omental wrap around the anastomosis is created in all patients.

Surgical procedure: MIE For the abdominal phase, the patient is placed in supine French position. The lesser omentum is opened and transected closely to the liver, until the left crus of the diaphragm is reached. Hereafter, the greater gastric curvature is dissected. An abdominal lymphadenectomy is performed including lymph nodes surrounding the hepatoduodenal ligament, the celiac trunk, along the left gastric and splenic artery and the lesser omental lymph nodes. The left gastric artery and vein are ligated with clips and transected at their origin. The gastric conduit is created at the level of the crow's foot using an endostapler For the thoracic phase, the patient is positioned in the left lateral decubitus position, tilted 45° towards the prone position (semi-prone). After selective desufflation of the right lung, the pulmonary ligament is divided. Hereafter, the parietal pleura is dissected at the anterior side of the esophagus from the diaphragm up to the azygos arch. The azygos vein is ligated with an Endostapler or clips. Dissection of the parietal pleura is continued above the azygos arch to establish dissection of the right paratracheal lymph nodes. At the posterior side of the esophagus, the parietal pleura is dissected cranially to caudally along the azygos vein, including the thoracic duct. The thoracic duct is clipped with a clips to prevent chylous leakage. The esophagus is resected en bloc with the surrounding mediastinal lymph nodes.

The gastric conduit is pulled up and the resection specimen is removed through a small incision (mini-thoracotomy) at the location of the trocar in the 6th intercostal space. Continuity is created using stapled esophago-gastrostomy, which is routinely oversewn with a V-Lock (Medtronic). An omental wrap around the anastomosis is created in all patients.

Statistical analysis All prospective data will be statistically analyzed by the use of the statistical software SPSS. All analyses were performed according to the intention-to-treat (ITT) principle. Results are presented as risk ratios with corresponding 95% confidence intervals (CI). To evaluate significance of differences between groups, the chi-squared test was used as appropriate for categorical variables and the student's T-test and non-parametric Mann-Whitney U-test for continuous variables.

Differences over time in quality of life and pain scores between and within treatment groups were assessed using linear mixed-effects models adjusted for the baseline value. Overall and progression-free survival curves were estimated with the Kaplan-Meier method and compared with the log-rank test. All reported P-values were two-sided. Significance level was set at 0.05.

The cost-effectiveness analysis will compare the mean costs and effects for both strategies and result in an incremental cost-effectiveness ratio. Uncertainty in the balance between costs and effects will be assessed with bootstrapping. A time horizon of 5 years will be applied, and costs and effects will be discounted according to Dutch guidelines.

If the baseline characteristics differ after randomization, i.e. there is a lack of balance in the confounding factors; this will be corrected using the multivariate analysis or by using a net benefit regression approach.

Interim-analysis There will be one interim-analysis. The stopping rule used for efficacy (i.e. better outcome for minimally invasive for the primary endpoint) is the Peto-approach, meaning a p-value <0.001. The trial will not be stopped for futility (i.e. no difference) and there is no there is no formal stopping rule for harm.

After every 50 patients, individualized patient description charts including safety parameters will be presented to the UGIRA Data Safety Monitoring Board (DSMB). The DSMB will discuss these in a plenary, telephone or online conference with the study coordinator and principal investigator present. The trial research group will discuss in a plenary session together with the DSMB the potential harm per patient and determine whether a relationship can be drawn between the surgical procedure and the adverse events. Consensus will be reached and the Institutional ethical board will be informed.