Surveillance vs. Endoscopic Therapy for Barrett's Esophagus With Low-grade Dysplasia

Methodology:

Patients with Barrett's esophagus and low grade dysplasia will be recruited in the multicenter trial. Patients will be randomized into endoscopic eradication therapy or endoscopic surveillance.

Subjects in the randomized control trial and observational cohort study, undergoing surveillance endoscopy will undergo surveillance biopsies in a 4-quadrant fashion every 1 cm throughout the extent of the Barrett's Esophagus using the Seattle biopsy protocol, along with targeted biopsies from any visible lesions. For incident low grade dysplasia (newly diagnosed low grade dysplasia - within 12 months of enrollment), surveillance endoscopies will be performed every 6 months for the first year and then annually until the end of the study period. For prevalent low grade dysplasia (diagnosed >1 year prior to enrollment), surveillance endoscopies will be performed annually until the end of the study period. The number of evaluations will depend on a subject's enrollment time with a maximum follow up period of 4years.

Subjects undergoing endoscopic eradication therapy will undergo radiofrequency ablation every 2-3 months until complete eradication of intestinal metaplasia (CE-IM) is achieved or 5 treatments have been delivered, whichever is first. After achieving CE-IM, surveillance endoscopy will performed every 6 months for the first year and annually thereafter until the end of the study period. Surveillance biopsies will be obtained using a standardized protocol.

Subjects will be contacted 48-72 hours and 30 days post procedure. All subjects will also receive follow-up phone calls on a semi-annual basis by a blinded central study coordinator.

Study Centers:

To maximize the generalizability of results, this randomized controlled trial will be conducted across different practice settings that include tertiary care centers, closed healthcare networks and large community practices at approximately 21 sites.

Anticipated Number of Participants:

680

• Randomized Control Trial: 530 subjects (265 per study arm)
• Observational: 150 subjects (Cohort 1, n=100 and Cohort 2 n=50)

Statistical Methodology:

Sample size and power calculations were performed for the primary endpoint using a time-to-event analysis. Estimates from available published data were used to approximate the expected progression rates in each arm of the trial. Based on previous clinical trials using similar methodology to confirm diagnosis of low grade dysplasia by expert pathology review, the team estimates 15% of patients with low grade dysplasia would progress to the composite primary endpoint in the surveillance arm compared to 6% in the endoscopic eradication therapy arm. T plan to accrue subjects for 3.5 years and follow them over time and record their time until progression to the primary endpoint or their censoring time if they do not progress. Follow-up observation will continue for approximately 1 year after the last subject is enrolled. Using this term of follow-up and assuming an exponential survival curve in each group and one interim analysis for efficacy and futility, 213 subjects are needed for analysis in each group to achieve 80% power using a two-sided 0.05 alpha level. Thus, accounting for a 10% non-adherence rate (attrition, subject cross-over) the team plans to enroll and randomize a total of 530 subjects (265 per study arm) who meet the eligibility criteria. A conservative rate of progression has been utilized for this sample size calculation given the significant heterogeneity in progression rates in the published literature. Recognizing that sample size estimation is based on assumptions and if the assumed event rate is lower than expected, there may be a decrease in power. To reduce the likelihood of an underpowered study due to incorrect assumptions, it is proposed to conduct a blinded sample size re-estimation once approximately 40% of the required events are reached.

All randomized subjects who are not identified at the index endoscopy with high grade dysplasia or post-endoscopy esophageal adenocarcinoma are defined as the intention to treat (ITT) population. The time to progression will be calculated from the time of randomization until the endoscopy date on which high grade dysplasia/mucosal post-endoscopy esophageal adenocarcinoma/invasive post-endoscopy esophageal adenocarcinoma is detected. If progression never occurs then the total time the subject is followed will be used as a censoring time. Time until censoring or progression to the primary endpoint of surveillance versus endoscopic eradication therapy will be compared by a log-rank test if the proportional hazards assumption is not violated. The continuous measures of Barrett's esophagus length will be included in the primary model as an independent variable. For all pre-specified analyses, a final two-sided p<0.05 will indicate statistical significance. This study is powered to test the primary hypothesis. However, it also offers the opportunity to conduct several analyses addressing other important patient outcomes. Analyses will be conducted to identify risk factors of progression, as well as factors associated with subsequent absence of low grade dysplasia during follow-up using logistic regression analyses. Potential confounding baseline variables, such as demographics, presence of visible lesions, confirmed low grade dysplasia, multifocal low grade dysplasia, and center differences, will be examined.