Incurable oesophageal cancer remains a global problem and in South Africa the vast majority of patients with oesophageal cancer have advanced disease at first presentation and are not curable. Likely the most distressing symptom of advanced cancer in the oesophagus is dysphagia, which is the inability to swallow solids and later also liquids. This is successfully addressed in most cases by the placement of a stent in the oesophagus which opens the area of obstruction. When placed in the lower oesophagus, one of the major drawbacks of these stents is that they disrupt the anti-reflux mechanism of the oesophago-gastric junction, which can result in severe acid reflux, severely impacting the quality of life of the patient. To address this problem, a range of approved anti-reflux stents have been developed and tested in numerous trials. To date, the evidence is conflicting and there is insufficient current evidence to support the routine use of these stents. However, the trials are not all similar in how the acid reflux was measured or what type of stent was used. Furthermore, the use of anti-reflux medication, such as proton pump inhibitors, which may help reduce reflux, are not standardised across the trials and make further conclusions about these stents difficult to interpret.

No data from Sub-Saharan Africa on the use of anti-reflux stents in these patients is available. South Africa faces a large burden of incurable oesophageal cancer and improving the quality of life of these patients is of paramount importance. This randomised controlled trial aims to investigate whether anti-reflux stents do indeed reduce acid reflux in patients with incurable oesophageal cancer compared to conventional oesophageal stents that do not have such an anti-reflux mechanism. Reflux will be measured using patient questionnaires about reflux, and other quality of life parameters, and will also be objectively measured using oesophageal scintigraphy, which has not been used in previous similar trials.

Oesophageal cancer is an aggressive condition, resulting in the vast majority of patients having evidence of locally invasive, irresectable disease or distant metastases at the time of presentation. Overall survival remains poor. Management in the South African setting is plagued by late presentation of these patients, with less than 5% being eligible for curative treatment and median survival from the time of diagnosis being only 15 weeks, while those who present with complete obstruction having a median survival of only 75 days (10.7 weeks). Treatment of these patients in the South African setting remains predominantly palliative. The most common and debilitating symptom of advanced oesophageal malignancy is progressive dysphagia, which can be addressed by the endoscopic placement of self-expanding metal stents. The major drawback of stenting tumours in the lower oesophagus or oesophagogastric junction (OGJ), is the associated gastro-oesophageal reflux (GOR) resulting from the stent crossing the lower oesophageal sphincter and essentially negating the native anti-reflux mechanism. Significant reflux is the most common complaint worsening quality of life after stent placement in these patients and can be as high as 100% in some series. Prescribing routine proton pump inhibitors (PPIs) or placing stents with built-in anti-reflux mechanisms are methods aimed at reducing this symptomatic reflux.

Theoretically, oesophageal stents containing an anti-reflux valve should provide a physical barrier to prevent gastric content (which may be acidic or non-acidic) refluxing into the oesophagus, but whether this results in decreased rates of GOR in reality is somewhat controversial. To date, a number of trials have compared a range of anti-reflux oesophageal stents to conventional oesophageal stents and although there have been some conflicting results, a systematic review and meta-analysis in 2019 concludes that GOR is not significantly reduced by the use of anti-reflux stents. However, there are a number of factors that must be mentioned before this conclusion can be applied to dictate clinical practice. Firstly, the included trials all have reasonably small participant numbers, with 65 patients being the highest number of patients enrolled in any of these trials. In fact, the authors conclude that the meta-analysis is underpowered. Furthermore, the type of anti-reflux stent used varies with almost every trial and may well influence efficacy of reducing GOR. Anti-reflux medical therapy such as the use of proton pump inhibitors (PPIs) also varied greatly amongst the studies. Some prescribed PPIs only to the conventional stent group, others did not use PPIs in either group, while the rest did not mention whether PPIs were routinely given or not. This could possibly influence symptomatic reflux and act as a significant confounding factor.

The measurement of GOR in the trials assessed in this meta-analysis shows significant heterogeneity, with some studies using patient questionnaires (some of these assess quality of life in general and do not specifically focus on reflux symptoms), others use contrast oesophagography or functional 24-hour pH monitoring. These additional factors make the results of this underpowered meta-analysis difficult to interpret. Since then, a further randomised controlled trial (RCT) was conducted by Dua et al. This included a total of 60 patients, comparing a novel tricuspid-shaped valve anti-reflux stent (30 patients) to conventional stenting (30 patients). Importantly, this trial was a non-inferiority trial to assess safety and efficacy at improving dysphagia for the new stent. Assessment of GOR was a secondary outcome and although reflux rates favoured the new anti-reflux stent, this did not reach statistical significance. The current level I and II evidence on reducing GOR with anti-reflux stents is thus not definitive and leaves the topic unresolved.

While research in high income countries is focused on the management of early oesophageal malignancies, this is not appropriate in the South African setting where the vast majority of patients are irresectable at initial presentation. Local research is significantly limited and there is a paucity of data from South Africa, and Africa as a whole, as regards the palliative management of malignant oesophageal dysphagia. Specific evidence on the use of anti-reflux stents is absent. Further research is thus invaluable in assessing if the palliative care of these patients can be improved by using anti-reflux stents.

This prospective randomised controlled trial aims to compare the incidence of symptomatic volume GOR after the use of anti-reflux oesophageal covered metal stents versus conventional oesophageal covered metal stents for lower oesophageal malignant strictures in a South African tertiary referral centre with a high rate of palliative stenting for advanced oesophageal carcinoma. Reflux will be assessed subjectively by the administration of patient questionnaires aimed at identifying severity of acid reflux, but also the degree of dysphagia, pain and coughing before and after stent placement. Reflux will also be assessed objectively by using oesophageal scintigraphy performed shortly after stent insertion. Scintigraphy has not previously been used to measure GOR in these patients and is chosen for its accuracy and non-invasive nature (compared to, for example, pH monitoring which requires the placement of an uncomfortable nasal probe for 24 hours and is considered inappropriate in this cohort where the main focus lies on quality of life).

Inclusion Criteria:

• Adult patients - 18 years of age or older
• Informed consent obtained from the patient after oral and written explanation of the trial
• Histologically confirmed malignancy of the distal oesophagus or OGJ
• Obstructive or irresectable malignancy due to metastases, local tumour infiltration or poor performance status
• Once deployed, the distal end of the stent must have crossed the OGJ junction and be lying within the proximal stomach

Exclusion Criteria:

• Patient declining or unable to give informed consent, including inability to speak or understand either English, Afrikaans or isiXhosa (the three most commonly spoken local languages).
• Patient unable to comply with the follow-up protocol of the trial (e.g. does not have a contactable telephone number)
• Oesophageal cancers selected for curative treatment or irresectable oesophageal cancers selected for palliative chemoradiation, but not requiring oesophageal stenting
• Benign oesophageal pathology or extrinsic compression of the oesophagus from another cause
• Patients with oesophageal cancers where the stent does not cross the OGJ
• Pregnant patients
• Patient performance status precluding any intervention or sedation

Individual Participant Data (IPD) that underlie the results reported in the published article for this study, after deidentification, will be made available to any investigator or investigators who request the data for analyses to achieve specific aims of a proposed research study. The study and proposed use of the data must be approved by an independent review committee identified for this purpose. The Study Protocol and Informed Consent Form will also be made available on request. Data will be available immediately after publication of the study and for 5 years after this. Proposals should be directed to matthias.scriba@gmail.com. to gain access, data requestors will need to sign a data access agreement.

Data will be available immediately after publication of the study and for 5 years after this.

The study and proposed use of the data must be approved by an independent review committee identified for this purpose. Proposals should be directed to matthias.scriba@gmail.com. to gain access, data requestors will need to sign a data access agreement. Requests will be assessed by the study's Primary Investigator (PI) and the Study Sponsor, the University of Cape Town.

Participants will remain unaware of which type of stent they have received. Follow-up will be performed telephonically and the Outcomes Assessor will also remain blinded as to what type of stent the participant received