AIM-IBD: A Microbiome-Targeted Supplement in Mild-to-Moderate Ulcerative Colitis

Responder defined by all four criteria at Week 24: (1) rectal bleeding subscore = 0 on the Mayo 0-3 scale (lower = less bleeding; 0 = no blood seen); (2) stool frequency subscore ≤1 on the Mayo 0-3 scale (lower = fewer stools above the participant's normal) with at least a 1-point decrease from baseline; (3) fecal calprotectin ≤250 µg/g (lower = less mucosal inflammation), measured centrally; (4) no rescue therapy, prohibited escalation, UC-related hospitalization, colectomy, or discontinuation for lack of efficacy through Week 24. Participants not meeting any criterion, or with missing Week 24 data, are counted as non-responders (non-responder imputation). Higher proportion = greater treatment benefit. Analyzed in the intention-to-treat population using stratified Cochran-Mantel-Haenszel adjusted for randomization stratification factors, reporting absolute risk difference with Newcombe 95% CI.

Responder defined as Mayo endoscopic subscore ≤1 at Week 24 (scale 0-3; lower = less inflammation; 0 = normal, 1 = mild erythema/decreased vascular pattern, 2 = marked erythema/friability/erosions, 3 = spontaneous bleeding/ulceration). Week 24 assessment is by protocol-mandated flexible sigmoidoscopy of the most inflamed segment with standardized image/video capture, scored by two independent blinded readers; discordant scores or scores crossing the ≤1 vs >1 threshold are adjudicated by a third blinded reader (adjudicator score = trial-of-record). Participants meeting any treatment-failure criterion before Week 24, or not undergoing the Week 24 endoscopy without documented medical exception, are counted as non-responders. A completer-only sensitivity analysis is prespecified. Higher proportion = greater treatment benefit. Analyzed in the intention-to-treat population using stratified Cochran-Mantel-Haenszel; effect estimate is absolute risk difference (Newcombe 95% CI).

Responder defined as fecal calprotectin ≤150 µg/g at Week 24, measured centrally using a validated assay. Fecal calprotectin is a stool neutrophil-derived biomarker of intestinal inflammation: healthy adults typically <50 µg/g; well-controlled UC <150-250 µg/g; active disease often several hundred to several thousand µg/g. The 150 µg/g threshold is stricter than the 250 µg/g threshold used in the primary composite endpoint and corresponds to the deep biochemical remission target in STRIDE-II treat-to-target guidance. Participants meeting any treatment-failure criterion before Week 24, or with missing Week 24 calprotectin, are counted as non-responders (non-responder imputation). A completer-only sensitivity analysis is prespecified. Higher proportion = greater treatment benefit. Analyzed in the intention-to-treat population using stratified Cochran-Mantel-Haenszel adjusted for randomization stratification factors; effect estimate is absolute risk difference (Newcombe 95% CI).

Continuous endpoint quantifying treatment effect on intestinal inflammation. Fecal calprotectin is a stool neutrophil-derived biomarker measured centrally using a validated assay at baseline and Week 24; healthy adults typically <50 µg/g, well-controlled UC <150-250 µg/g, active disease often several hundred to several thousand µg/g, with no fixed upper limit. Lower values indicate less mucosal inflammation. Values are log-transformed (natural log, after substituting half the lower limit of quantification for any below-LLOQ result), and the treatment effect is summarized as the geometric mean ratio of active vs placebo at Week 24 with 95% confidence interval, derived from a mixed model for repeated measures with treatment, visit, treatment-by-visit interaction, log baseline calprotectin, and randomization stratification factors as fixed effects. A geometric mean ratio below 1 indicates a greater reduction in fecal calprotectin in the active arm relative to placebo.

Continuous endpoint quantifying early treatment effect on intestinal inflammation, prespecified as an interim biomarker readout midway through the 24-week treatment period. Fecal calprotectin is a stool neutrophil-derived biomarker measured centrally using a validated assay at baseline and Week 12; healthy adults typically <50 µg/g, well-controlled UC <150-250 µg/g, active disease often several hundred to several thousand µg/g. Lower values indicate less mucosal inflammation. Values are log-transformed (natural log, after substituting half the lower limit of quantification for any below-LLOQ result) and the treatment effect is summarized as the geometric mean ratio of active vs placebo at Week 12 with 95% confidence interval, derived from a mixed model for repeated measures with treatment, visit, treatment-by-visit interaction, log baseline calprotectin, and randomization stratification factors as fixed effects. A ratio below 1 indicates greater reduction in the active arm vs placebo.

Responder defined by meeting both: (1) the primary composite clinical-plus-biochemical remission endpoint at Week 24 (rectal bleeding subscore = 0; stool frequency subscore ≤1 with at least a 1-point decrease from baseline; fecal calprotectin ≤250 µg/g; and no rescue therapy, prohibited escalation, UC-related hospitalization, colectomy, or discontinuation for lack of efficacy through Week 24); and (2) no use of systemic or rectal corticosteroids (including budesonide-class) at any time between randomization and the Week 24 assessment. Corticosteroid exposure is captured from concomitant medication records and reconciled with rescue therapy documentation. Participants not meeting either criterion, or with missing Week 24 data, are counted as non-responders (non-responder imputation). Higher proportion = greater treatment benefit. Analyzed in the intention-to-treat population using stratified Cochran-Mantel-Haenszel; effect estimate is absolute risk difference (Newcombe 95% CI).

Continuous endpoint quantifying treatment effect on clinical disease activity. The partial Mayo score is the sum of three Mayo subscores - stool frequency (0-3, lower = fewer stools above the participant's normal), rectal bleeding (0-3, lower = less bleeding; 0 = no blood seen), and physician global assessment (0-3, lower = less overall disease severity in the physician's judgment) - yielding a total score from 0 to 9, with lower scores indicating less disease activity. Assessed at baseline and Week 24. Treatment effect is summarized as the least-squares mean difference between active and placebo in change from baseline to Week 24 with 95% confidence interval, derived from a mixed model for repeated measures with treatment, visit, treatment-by-visit interaction, baseline partial Mayo, and randomization stratification factors as fixed effects. A more negative change indicates greater reduction in disease activity; a between-arm difference favoring active = greater treatment benefit.

Continuous endpoint quantifying treatment effect on disease-specific quality of life. The Inflammatory Bowel Disease Questionnaire-32 (IBDQ-32) is a 32-item validated patient-reported instrument covering four domains: bowel symptoms, systemic symptoms, emotional function, and social function. Each item is rated 1-7 (1 = worst, 7 = best); the total score ranges from 32 to 224, with higher scores indicating better quality of life. A change of 16 points or more is generally considered clinically meaningful. A locally validated Turkish translation is administered at baseline and Week 24. Treatment effect is summarized as the least-squares mean difference between active and placebo in change from baseline to Week 24 with 95% confidence interval, derived from a mixed model for repeated measures with treatment, visit, treatment-by-visit interaction, baseline IBDQ-32, and randomization stratification factors as fixed effects. A larger positive change in the active arm = greater benefit.

Continuous endpoint quantifying treatment effect on systemic inflammation. C-reactive protein (CRP) is a serum acute-phase reactant of hepatic origin, measured in mg/L by the central laboratory using a high-sensitivity assay at baseline and Week 24. Healthy adults typically have CRP <3 mg/L; lower values indicate less systemic inflammation. Note that CRP can be normal in a sizeable subset of UC patients with active mucosal disease, particularly those with distal/left-sided disease. Values are log-transformed and the treatment effect is summarized as the geometric mean ratio of active vs placebo at Week 24 with 95% confidence interval, derived from a mixed model for repeated measures adjusting for log baseline CRP and randomization stratification factors. A ratio below 1 indicates greater reduction with active treatment.

Safety endpoint summarizing the cumulative incidence and exposure-adjusted rates of adverse events over the treatment period. A treatment-emergent adverse event (TEAE) is any adverse event with onset, or worsening, on or after the first dose of study product through 4 weeks after the last dose. Reported by treatment arm: number and proportion of participants with any TEAE; any serious adverse event (SAE); any TEAE leading to study product discontinuation; and any adverse event of special interest (AESI). AESI: new-onset Clostridioides difficile infection; serious enteric or opportunistic infection; UC worsening requiring hospitalization; colectomy; hepatobiliary events; severe allergic or anaphylactic reactions; gastrointestinal perforation; death. Coded with MedDRA (version current at database lock); severity graded per CTCAE v5.0; relatedness assessed by the investigator. Higher numbers and proportions indicate more frequent events.

Exploratory endpoint characterizing treatment-associated shifts in gut microbial communities. Stool samples are collected at baseline, Week 12, and Week 24 under a harmonized SOP and processed centrally. Sequencing is by 16S rRNA gene profiling and/or shotgun metagenomics (final choice locked in the microbiome analysis plan before database lock), with negative controls, mock communities, and batch randomization. Outputs include alpha diversity (Shannon index, Faith's phylogenetic diversity; higher = more diverse), beta diversity (Aitchison and weighted/unweighted UniFrac distances; quantifying between-sample compositional difference), differential abundance of taxa using compositionally appropriate transformations (CLR/ILR) with Benjamini-Hochberg false discovery rate control at q<0.05, and treatment-associated taxonomic and functional shifts. As an exploratory endpoint, no formal hypothesis testing within the confirmatory hierarchy is performed.