The primary endpoint in this study was the change form baseline in the percentage of aberrant intestinal intraepithelial lymphocytes (IELs) with respect to total IELs, as assessed by flow cytometry (Immunological Response 1).

Intraepithelial lymphocytes (IELS) are white blood cells interspersed between epithelial cells of the small and large intestine where they function to preserve the integrity of the mucosal barrier by protecting the epithelium against pathogen or immune-induced pathology. In refractory coeliac disease type 2, aberrant intraepithelial lymphocytes make up 20% or more of total intraepithelial lymphocytes. Aberrant IELs were defined by flow cytometry as surface cluster of differentiation (CD)3-negative, intracellular CD3-positive IELs (sCD3-, icCD3+).

The per protocol (PP) population included participants who received study drug and provided evaluable data for efficacy analysis, excluded non-evaluable participants and those with major protocol deviations. Participants with atypical RCD-II were also excluded from analyses of Immunological Response 1.

Percent change from baseline in the percentage of aberrant intestinal IELs with respect to intestinal epithelial cells (Immunological Response 2) is a composite endpoint calculated by multiplying the percent of aberrant IEL versus total IELs (per flow cytometry) by the percent of total IEL versus intestinal epithelial cells as assessed by immunohistochemistry.

The per protocol (PP) population included participants who received study drug and provided evaluable data for efficacy analysis, excluded non-evaluable participants and those with major protocol deviations. Participants with atypical RCD-II were also excluded from analyses of Immunological Response 2.

Villi are the small fingerlike projections that line the small intestine and promote nutrient absorption and are often shortened in patients with celiac disease. Crypts are grooves between the villi that are often elongated in patients with celiac disease. A decreased VH:CD ratio indicates worsening disease and increases in the VH:CD ratio indicate an improvement in the histology of intestinal mucosa (Histological Response).

Small bowel biopsies were performed at baseline and week 12; histological assessments were performed by a blinded central pathologist.

Per protocol population

The Marsh classification system describes the stages of damage in the small intestine as seen under a microscope, with possible values of 0, 1, 2, 3a, 3b, or 3c. A score of 0 (best score) indicates that the intestinal lining is normal and celiac disease highly unlikely, a score of 3c (worst score) indicates increased intraepithelial lymphocytes, increased crypt hyperplasia and complete villi atrophy.

Improvement is defined as a lower grade on the Marsh score scale compared to baseline.

Per protocol population

Small bowel biopsies were performed at baseline and week 12; histological assessments were performed by a blinded central pathologist. The total IEL count is the density of IELs vs intestinal epithelial cells measured by immunohistochemistry.

The per protocol population

Participants were asked to record every bowel movement during the study using an electronic diary. If no bowel movements were experienced by the participant on any given day, the participant was required to document this in the diary.

The intent-to-treat population consisted of all randomized participants who had received at least one dose of the study drug.

The Bristol Stool Form Scale (BSFS) is a pictorial aid to help participants identify the shape and consistency of their bowel movements. Participants were asked to complete this form daily using an electronic diary at the time of each bowel movement. The BSFS categorizes bowel movements into 7 types, from Type 1 (separate hard lumps, like nuts; hard to pass) to Type 7 (watery, no solid pieces, entirely liquid).

Diarrhoea was defined at least one BSFS score >= 6 for the given week.

Intent-to-treat population

The GSRS is a 15-question 7-scale questionnaire used to assess 5 dimensions of gastrointestinal syndromes: diarrhea, indigestion, constipation, abdominal pain and reflux. Questions are scored between 1 (no discomfort at all) and 7 (very severe discomfort).

The total GSRS score is calculated as the sum of the scores of all 15 questions, and ranges from 15 (no discomfort at all) to 105 (very severe discomfort in all 5 dimensions of gastrointestinal syndromes).

Intent-to-treat population with available data.

The CeD-GSRS score is derived from a subset of 10 questions from the GSRS questionnaire (questions 1, 4-9, 11, 12 and 14), which are each assessed on a scale of 1 (no discomfort at all) to 7 (very severe discomfort).

The total CeD-GSRS score ranges from 10 (no discomfort at all) to 70 (very severe discomfort in all celiac syndromes).

Intent-to-treat population with available data.