MR1263 - BOSS - Barrett's Oesophagus Surveillance Study

BOSS is a randomised controlled trial, seeking to determine whether patients diagnosed with Barrett's Oesophagus, a pre-malignant condition, should be offered prospective surveillance by endoscopy every two years, as opposed to endoscopy only when prompted by symptoms. Consenting participants will be randomised to being offered surveillance endoscopy every two years, or to no scheduled endoscopy. Participants in both arms will have access to endoscopy whenever symptoms warrant it. Participants will be followed up for 10 years after recruitment.

Investigators at hospital sites identify potential participants who have clinic appointments; they may be newly-diagnosed with Barrett's Oesophagus, confirmed by endoscopy and histology, or have an existing diagnosis. Patients will receive the information sheet, and a discussion will be held in clinic (in rare cases on the telephone). Those who consent to BOSS will be randomised to either two-yearly invitations to endoscopy, or to no pre-planned endoscopies. In both arms, participants who present to their GP or hospital with symptoms suggesting possible progression will be eligible for endoscopy at the investigator's discretion. Participants are asked to complete quality-of-life questionnaires every two years, or at each endoscopy if this is more frequent. In the non-surveillance arm, these questionnaires are sent by the trial office in Gloucester.

The protocol does not prescribe how patients who progress to high grade dysplasia or cancer should be treated; in this event, surveillance ceases, but treatment continues according to local practice. Follow-up in the trial will continue until 10 years since the last participant was recruited. Data on frequency and complications of endoscopy, and on hospital admissions, will be collected by sites. Data collection will be complete by the end of 2021, and the report written by September 2022. Publication in a high-profile journal will then follow.

The primary notification of study endpoints (deaths and progressions to cancer) will come from study sites. However, a proportion of participants will be lost to follow-up. It is essential that follow-up be as complete as possible, as BOSS is likely to be the only randomised trial of surveillance endoscopy undertaken worldwide, and very considerable resources have been committed to it. This is the reason for registering all participants with MRIS/ISD. (Some participants choose not to be randomised, but have allowed their baseline data to be forwarded to the trial office anonymously, so that their characteristics can be compared with those who were randomised. These "data-only" participants will not be registered.)

Event notifications will be received by the trial office in Gloucester, but statistical analysis of all trial data will be undertaken by the Centre for Statistics in Medicine at the University of Oxford. All data received will be anonymised before transfer to Oxford. Health economic analysis, also of anonymised data, will be undertaken by the University of Glasgow.

The primary endpoint is a comparison of overall survival between the two trial arms. Secondary endpoints include progression to oesophageal cancer, and the incidence of all cancers. The frequency of endoscopy, incidence of complications, and the stage of oesophageal cancer at diagnosis will also be compared between trial arms. The cost-effectiveness of surveillance will be estimated. The trial results are expected to contribute greatly to future NHS and international policy on surveillance for Barrett's Oesophagus.

There are 5000 new cases of oesophageal adenocarcinoma (OA) in the UK per year, and the incidence is rising. If diagnosed only at an advanced stage, the outlook is poor, with 10% survival at 5 years. OA is preceded by metaplastic changes in the lining of the oesophagus, engendered by long-term reflux of stomach acids. These changes are termed Barrett's Oesophagus and are detectable by endoscopy following referral for reflux. The conversion rate of Barrett's Oesophagus to OA in the UK is estimated at 1% per year.

UK and US guidelines recommend endoscopic surveillance of Barrett's every 2-3 years, but the evidence is acknowledged to be weak, with no randomised trials. There are observational data and a small non-randomised comparison suggesting that surveillance detects OA at an earlier stage.

Surveillance endoscopy should not be undertaken lightly, as it is costly and provokes patient anxiety and discomfort, minor adverse events in 1-10% of cases, and an oesophageal tear in 0.03%. The BOSS trial has been funded to weigh up the costs and benefits of surveillance in as thorough a way as possible, and is likely to remain the only substantial randomised controlled trial in the field.

Not stated in the register.