Trial design
The PYNI-GAREREO trial is designed as an open randomized, single-center superiority trial. Patients will be randomly allocated to undergo gastric remnant reconstruction with mH-P or NI in parallel groups.
Methods: Participants, interventions and outcomes
Study setting
The PYNI-GAREREO trial will be conducted in Teine Keijinkai Hospital as a single-center, two-arm, open-label, randomized phase II superiority trial. Figure 1 shows the schedule of enrollment, interventions, and assessments. The SPIRIT reporting guidelines were used for this study protocol (10). The SPIRIT checklist is provided as Additional file 1.
Eligibility criteria
Population
All patients with resectable esophageal carcinoma or esophagogastric junction cancer will be screened for eligibility by esophageal surgeons.
Inclusion criteria
The patients will undergo MIE and gastric remnant reconstruction for radical treatment of esophageal cancer or esophagogastric cancer. The inclusion criteria are as follows.
(1) Reconstruction by the posterior sternal route
(2) No previous laparotomy with high degree of adhesion (e.g., traumatic incision for trauma surgery, duodenal ulcer perforation, gastrectomy)
(3) Age of ≥20 years at the time of providing informed consent
(4) Eastern Cooperative Oncology Group performance status of 0 or 1
(5) Treatment by thoracoscopic or robot-assisted thoracic MIE
(6) Abdominal operation by manual laparoscopy or laparoscopy
(7) Anastomosis by the cervical anastomosis technique
(8) Full understanding of the study and voluntary provision of written consent to participate in the study
Two esophageal surgeons will perform or assist all interventions.
Exclusion criteria
(1) Swallowing problems and poor oral intake associated with swallowing dysfunction
(2) Weight loss of ≥20% within 6 months before surgery
(3) Other types of active cancer or diseases that affect the nutritional status
(4) Gastric cancer requiring simultaneous resection
(5) Clinical condition inappropriate for participation in this study as judged by patient’s physician
The study will be conducted in Teine Keijinkai Hospital and will involve two esophageal surgeons (YK and NO) who are competent in performing both techniques.
Informed consent
Operators and clinical research coordinators (CRCs) will obtain informed consent or assent from potential trial participants. Operators will explain the procedure. CRCs will supportively explain the concept of the trial.
Randomization
After confirmation of the eligibility criteria, registration will be completed by a web-based system to the UMIN Clinical Trials Registry system. Patients will be randomized to the intervention group or control group in a 1:1 ratio using a computerized randomization tool (UMIN INDICE cloud (11)) with the minimization method, balancing the arms according to age, clinical TNM stage, and change in preoperative body weight during the most recent 6 months before surgery. The two arms will be the mH-P arm (intervention) and NI arm (control). Because this trial will involve different surgical techniques, complete blinding of the treating surgeons and medical staff is not feasible. The surgeon will know which operation has been performed, and the medical staff can determine whether the patients have undergone pyloroplasty or not by reading the official medical records. However, a study team member will assess blinded medical reports regarding mH-P or NI to ensure blinded assessment of the primary outcome parameter.
Pyloroplasty procedure
For gastric remnant reconstruction, mH-P either will or will not be performed according to the randomization result. mH-P will be performed by first longitudinally cutting all gastric wall layers to a length of 3 cm at 1.5 cm above and below the pyloric ring; the surgical wound will then be closed using seven to nine absorbable monofilament sutures (4-0 PDS®) in a horizontal Gambee pattern. Intraoperative botulinum injection and application of a pyloric ring finger bougie will be prohibited during the trial.
Decompression tube placement and jejunostomy
We will routinely perform decompression tube placement and jejunostomy of the gastric remnant. After mobilization of the stomach, we will cut the gastric serosa and muscular layer. We will then cut the point of the mucosa layer and insert two tubes: the decompression tube (inserted cranially) and the enteral tube (inserted caudally toward the jejunum). The two tubes will be placed to the gastric remnant wall by the Witzel procedure. Finally, after the reconstruction and cervical anastomosis, we will perform abdominal wall plasty of the gastric remnant with placement of tubes.
Criteria for discontinuing or modifying allocated interventions
If non-inferiority of the mH-P group to the NI group for the primary endpoint is demonstrated, and even if superiority is demonstrated in the interim, the study will be stopped (active discontinuation). In such a case, the need for discontinuation of the trial will be comprehensively considered without being restricted by the statistical judgment of the trial or other factors.
Prohibited procedures during the trial
Simultaneous resection for gastric cancer, other pyloroplasty procedures (e.g., finger bougie technique, Heineke–Mikulicz strictureplasty, and botulinum toxin injection), and Roux-en-Y reconstruction will be prohibited during the trial.
Provisions for post-trial care
In the event that a research participant suffers health damage as a result of the conduct of the research, the physician in charge of the research will provide appropriate treatment and investigate the cause of the damage. In such cases, if treatment or examination becomes necessary, it will be conducted within the scope of the research participant’s normal insurance treatment.
Outcomes
The primary endpoints of oral intake and change in nutritional status will be checked during the perioperative hospitalization period and at 1, 3, 6, 9, and 12 months after surgery.
Oral intake will be calculated as a percentage of total energy expenditure, which will calculated based on the weight and activity coefficient of each patient, and data will be collected at each time point. The nutritional status of the patient will be monitored by weight loss over time using the weight immediately before surgery as a control. In addition, body composition will be analyzed by a bioimpedance analyzer (InBody S10®). The analyzer determines the total body fat mass (kg) and skeletal muscle index (kg/m2). Additionally, blood tests will be performed at fixed points to measure nutritional evaluation indices (albumin, pre-albumin, cholesterol, total protein, lymphocyte count, and C-reactive protein).
Intraoperative factors such as the total blood loss, operation time, thoracic approach, subtotal gastric remnant or gastric tube reconstruction, and inclusion or exclusion of Kocher mobilization will also be analyzed.
To evaluate the degree of pyloric transit as a secondary endpoint, the amount and nature of decompressed gastric drainage and the pH and bilirubin level of the drainage will be measured on postoperative days 1, 3, 5, and 7. On day 7, the degree of pyloric drainage and the presence or absence of reflux will be assessed by oral contrast according to the standard clinical path. The blood and gastric drainage measurements will be taken in the laboratory, the data will be recorded in the medical record, and the sample will be discarded. The decompression tube will be clamped at 6:00 am to equalize the condition of the reconstructed elevated stomach, and contrast examination will be performed at 9:00 am on the same day. The contrast medium will be barium diluted to 60%.
The pyloric transit (an objective index) as evaluated by oral barium contrast will be scored according to the presence or absence of gastric contents in the reconstructed elevated stomach at the beginning of the examination (0: no, 1: yes), the pyloric transit time of the contrast medium (0: <1 second, 1: >1 second), and stagnation of the contrast medium in the stomach (0: completely disappears, 1: more than half remains, 2: all remains).
During this pyloric transit evaluation, whether the reconstructed gastric remnant has shifted to the thoracic cavity side will also be evaluated by the contrast examination. This examination may result in the performance of intraoperative thoracotomy via the mediastinum during the retrosternal route-creating procedure.
At 1, 3, 6, 9, and 12 months postoperatively, the above parameters and the amount of residue in the reconstructed raised stomach will be evaluated by computed tomography. At 6 and 12 months postoperatively, the amount of residue in the reconstructed raised stomach and the pyloric transit will be evaluated by endoscopy.
Postoperative gastrointestinal symptom questionnaires will be requested at the time of discharge, at the first outpatient visit after discharge, and at 3, 6, 9, and 12 months after surgery.
The short-term outcomes evaluated in this study will be postoperative complications, especially anastomotic leakage, and the length of hospital stay.
The routinely placed jejunostomy enteral tube will be removed within 3 months postoperatively in the outpatient clinic when the patient has been determined to have no risk of reduced oral intake. If the tube feeding is used for a longer period, the prolonged time will be one of the outcome parameters.
Long-term recurrence and the prognosis will also be followed up.
Patients will be blinded to which group they are in, but the physician who performs the surgery and performs the imaging and follow-up will not be blinded to whether pyloroplasty is performed. The dietitian, nurse, and data analyst who evaluate the nutritional status and gastrointestinal symptoms will be blinded to which group each patient is in.
Sample size
From the pilot date, the sample size estimation also corresponds to data from our own database containing data on body weight changes after esophageal resection and gastric remnant reconstruction via the retrosternal route, and this body weight change is clinically relevant. Based on the superiority trial design, alpha of 0.05, and power of 80%, 70 patients should be included in each treatment arm.
Recruitment
Surgeons and CRCs will check the inclusion and exclusion criteria 1 to 2 days before an eligible patient visits their outpatient clinic. Inclusion rate feedback will be provided every 6 months in a monitoring report. Completeness of case record form (CRF) data and adherence to the study protocol will be checked on a weekly basis by the coordinating investigator or CRC.
Data collection and management
A participant number will be generated upon each patient’s inclusion in the study, and this number will be used for further identification in the database. The participant number key is accessible by the coordinating investigator. Clinical data will be collected by the study coordinator or clinical research nurse and will be recorded in a good clinical practice-compliant digital CRF and database (UMIN (11)). All non-electronic items containing data will be kept in locked cabinets at the data coordinating centers.
The data will be able to be accessed by the research nurse and research physician. After the study has been completed, requests to access the dataset can be submitted to the project leader or principal and coordinating investigator. The completed CRFs will also be checked with the source data regarding the primary outcome parameter and important secondary outcome parameters.
Follow-up
For each participant, the study will start at randomization and the patient will be followed until 12 months after surgery. The primary outcome parameter will be evaluated every 3 months after surgery. During the 1 year of follow-up, data on readmission, functional results, and quality of life will be generated. Study visits will be scheduled to take place 4 weeks before surgery and 1, 3, 6, 9, and 12 months after surgery.
Statistical methods
The Mann–Whitney U-test will be used to compare continuous variables such as weight change, body composition, and blood test indices, which are the primary endpoints. For the secondary endpoints, Student’s t test (two-tailed) will be used to compare mean parameters, and the chi-square test will be used to compare bivariate variables such as the presence of perioperative complications. Survival will be compared by the log-rank test.
Further subgroup analysis will only be carried out in case of significant interaction effects. In the case of missing data, we will perform Poisson regression analyses with a robust covariance matrix estimator to adjust for covariates because it has been shown that complete case analysis with covariate adjustment and multiple imputation yield similar estimates in the event of outcome data that are missing at random (12).
Interim analyses
Two interim analyses will be performed during the course of the study to determine whether the primary objective of the study has been achieved. The first interim analysis will be performed during enrollment to determine whether it is appropriate to continue enrollment, and the second interim analysis will be performed early after the end of enrollment to determine whether to continue follow-up for the planned period. In either case, if it is determined that the primary objective of the study has been achieved, the study will be terminated and the results will be promptly published in conferences and papers.
The first interim analysis will be conducted using data from periodic monitoring when half of the expected number of enrolled patients have completed 1 year of follow-up. The second interim analysis will be conducted in conjunction with periodic monitoring at a time deemed appropriate after consultation between the data center and the CRC, around the time when enrollment is completed, and protocol treatment is completed for all enrolled patients. In principle, enrollment will not be stopped during the first interim analysis. The decision criteria based on the results of the interim analysis of this study will be as follows. If non-inferiority of the mH-P group to the NI group for the primary endpoint is not demonstrated, or if non-inferiority is demonstrated but superiority is not demonstrated, the study will be continued in either case. If non-inferiority of the primary endpoint of mH-P over NI is demonstrated and superiority is also demonstrated, the study will be terminated (effective termination). If the primary endpoint of the mH-P group is lower than that of the NI group, the need for discontinuation of the study will be comprehensively considered without being restricted by statistical judgment such as tests (invalid discontinuation).
Oversight and monitoring
An independent data and safety monitoring committee will evaluate the progress of the trial and examine safety variables. The committee will consist of a surgeon, CRCs, and a statistician. Individualized patient description charts including safety parameters will be presented to the committee, including one table comprising these endpoints in blinded groups for every 6 months. The main safety parameters are all serious adverse events (mortality, multiple organ failure, anastomotic leakage, pulmonary complications, cardiovascular complications, reinterventions, and reoperation). After the investigators have presented the data, the members of the committee will discuss these results in the absence of the investigators and will then advise them. Possible options will include continuing the trial, performing an interim analysis, adjusting the trial’s design, and discontinuing the trial. Discontinuation will be advised if the committee concludes that the results would convince a broad range of clinicians that one trial arm is inferior or if safety is compromised in one arm. If the committee advises adjustment of the trial’s design, performance of an interim analysis, or discontinuation of the trial, the responsible medical ethical committee will also be notified. Serious adverse events will be reported to the chairman of the hospital. The PYNI-GAREREO trial will be monitored according to the Japan ethical guidelines for medical and health research involving human subjects. This is a low-risk study because both interventions being investigated are considered to be standard care in Japan.
The first interim analysis will be conducted after half of the planned number of patients are enrolled, and the second interim analysis will be conducted immediately before the planned patient accrual is completed. The data and safety monitoring committee will review the interim analysis reports independently from the investigators and statistician. In-house monitoring will be performed every 6 months by a data center to evaluate and improve the study progress, data integrity, and patient safety.