Statistical Analysis

Statistical analyses and reporting of this trial will be conducted with primary analyses based on the intention-to-treat approach. The full analysis set will include all randomized subjects administered at least one procedure of the investigational treatment. For baseline variables, we will generate summary statistics with proportions and frequencies for categorical variables, and means and standard deviations for continuous data. Statistical data relevant to feasibility will be presented descriptively. For primary and secondary clinical outcome analyses, we will analyze mean changes from baseline with a restricted maximum likelihood-based repeated measures approach. The mixed model for repeated measures analyses will include the fixed and categorical effects of treatment, visit, and the treatment×visit interaction. We will employ Kenward-Roger approximation to estimate the degrees of freedom of the denominator. We will not conduct any adjustment for multiple testing of secondary outcomes because of the exploratory nature of the study. Imputation will not be performed for missing values because mixed models can deal with missing data by maximum likelihood. All comparisons are planned, and all P values are two-sided. A 5% significance level will be set for all statistical analyses. All statistical analyses will be conducted using Stata version 16 (Stata Corp).