Statistical Analysis Plan the intention-to-treat (ITT) population (all randomized participants analyzed according to the randomized assignment)-the primary analysis set; the per-protocol (PP) population (participants with ≥80% treatment compliance and no major protocol violations)-used for confirmatory efficacy consistency testing; and the safety analysis (SA) population (all randomized participants exposed to the investigational product or placebo with post-intervention safety data)-used for all safety evaluations.
Primary Outcome Analysis Due to the asymptomatic nature of colorectal adenoma recurrence, which can only be detected at scheduled annual colonoscopic examinations, the time to recurrence is considered interval-censored data (i.e., the exact recurrence time is unknown, only that it occurred within a 12-month window between two consecutive colonoscopies). Given this data characteristic, the 3-year cumulative CRA recurrence rate will be estimated using the life table method, which divides follow-up time into 12-month intervals (0-12, 12-24, 24-36 months). Between-group comparisons of recurrence-free survival will be performed using the stratified log-rank test (stratified by participating center). Hazard ratios (HR) and corresponding 95% confidence intervals (CIs) will be estimated using a stratified Cox proportional hazards model. To ensure the robustness of the findings, sensitivity analyses using the Turnbull algorithm (an extension of the Kaplan-Meier method for interval-censored data) will also be conducted.