Calculating the LEmann IndeX Using InTEstiNal Ultrasound

EXTENT will be a multicenter, cross-sectional study including 102 Crohn's disease (CD) patients, stratified by known or suspected CD location (small bowel and colon) and by disease duration (< 2 years, ≥ 2 years and < 10 years, and ≥ 10 years). The investigators will recruit patients with an established diagnosis of CD, with no restriction per disease activity, surgical history or CD medication.

Enrolled participants will undergo baseline blood work, stool studies, and both MRE and IUS.

Individuals with colonic involvement will also undergo colonoscopy. Patients will also answer to the IBD disability index questionnaire to assess the patients' functional status. All examinations performed will be standard of care, needed for patient monitoring, and the results will be used for the purpose of this study. Clinical data, imaging and colonoscopy assessments will be collected from participants prospectively over a period of 8 weeks. IUS and MRE data will be centrally read by three central readers blinded to the local reader grading of lesions.

The primary objective is to demonstrate that IUS can replace MRE to evaluate small bowel global damage and colon global damage (excluding rectum) in Crohn's disease patients with small bowel and/or colonic involvement.

Additive objectives concerning IUS versus MRE are:

• To demonstrate that IUS can replace MRE + colonoscopy to evaluate colon global damage (excluding rectum).
• To demonstrate that IUS can replace MRE in assessing segmental damage in patients with CD (terminal ileum).
• To demonstrate that IUS can replace MRE + colonoscopy in assessing segmental damage in patients with CD (all colonic segments, except rectum).
• To test if local investigators can replace central readers to evaluate small bowel, colon and global LI-IUS and LI-MRE.
• To quantify how much higher is IUS acceptability relatively to MRE and colonoscopy acceptabilities.

Inter-reader variation study aims:

• To demonstrate that it is possible for central readers to evaluate in a reliable way small bowel, colon and global LI-IUS scores.
• To confirm that it is possible for central readers to evaluate in a reliable way small bowel, colon and global LI-MRE scores.
• To compare the intra- and inter-reader reliability levels of small bowel, colon and global LI-IUS scores and of small bowel, colon and global LIMRE scores.
• To demonstrate that it is possible for central readers to re-evaluate in a reliable way small bowel, colon and global LI-IUS scores.
• To confirm that it is possible for central readers to re-evaluate in a reliable way small bowel, colon and global LI-MRE scores.

Other objectives are:

• Relationship between IBD disability and global, small bowel and colon damage as evaluated by IUS and by MRE.
• Comparison of the presence and distribution of stricturing and penetrating lesions using IUS and MRE, across severity levels.

The exploratory objectives are:

• To evaluate potential intestinal ultrasound parameters equivalent to wall hyperenhancement in MRE - used to define type 1 stricturing lesions.
• To evaluate potential intestinal ultrasound parameters equivalent to mural stratification in MRE - used to define type 2 stricturing lesions.
• To identify additional features of bowel wall damage on intestinal ultrasound.

Inclusion criteria:

• Adult patients (≥18 years old).
• Established diagnosis of CD, in line with international diagnostic criteria.
• Examinations needed as standard of care.
• Patients able to understand the information provided to them and to give written informed consent for the study.

Exclusion criteria:

• Pregnant women or patients willing the conceive.
• Patients unwilling or unable to provide informed, written consent.
• Uncertain CD diagnosis or Colitis indeterminata.
• Patients with disease limited to the upper gastrointestinal tract, rectum or perianal CD.
• Patients with a contraindication for MR, colonoscopy and/or ultrasound.
• Patients with allergy to gadolinium chelated or with Estimated Glomerular Filtration Rate <60 mL/min/1.73m2.

Intestinal Ultrasound:

IUS examinations will be performed by a gastroenterologist or radiologist using different US devices and low-frequency and mid to high-frequency probes from various manufacturers. IUS examinations will be done preferably in a non-fasting state and will consistently capture all small bowel (jejunum and ileum) and colonic segments (cecum, ascending colon, transverse colon, descending colon and sigmoid colon). All investigators in the study will be trained in LI calculation and on the specific lesions that should be assessed to calculate the LI (Appendix 4). Both probes (low-frequency and mid to highfrequency) should be used to assess IUS lesions for IUS-Lémann index. The affected small bowel or colonic segments will be graded for the worse type of stricturing or penetrating lesion. The researcher will record all intestinal segments and upload the cineloops to a central reading platform and indicate in real time the lesions observed in each affected segment. This will allow the investigators to test feasibility of use of LI-IUS in daily clinical practice.

If small bowel is involved, the extent of disease will be measured, and the number of 20 cm segments affected will be collected. If more than one pathological 20-cm small bowel segment is found, they will be labeled as SB1 (beginning in the ileocecal valve or ileocolic anastomosis), SB2, SB3, etc., preferably adding a body marker OR text for location (ex: LLQ, RUQ). A complete small bowel scan cineloop will be recorded to assess small bowel lesions and disease extension. All pathological small bowel segments identified during the examination need to be recorded in a longitudinal section to allow for disease extension measurement, with a cineloop of 5-10 seconds, preferably with a mid to high-frequency probe, if the complete segment can be visualized. Every pathological small bowel segment should also be recorded in cross-sectional, with a cineloop of 5-10 seconds. Every colonic segment should be recorded in both cross-sectional and longitudinal planes, with a cineloop of 5-10 seconds. For the purpose of the LI the cecum is separated from the ascending colon.

The cecum corresponds to the colonic segment distal to the ileocecal valve. For anatomical reference the cineloop should include, if possible, the terminal ileum, the ileocecal valve, and the ileocecal appendix (if accessible). In non-operated individuals, a cineloop starting in the terminal ileum at the right lower quadrant (including anatomical landmarks such as psoas muscle and/or iliac vessels) and following parallel lines along the abdomen (extending from the liver/stomach/ribs to the bladder/iliac vessels) will be recorded. For post-operative patients (small bowel resection or ileocolonic resection) a cineloop starting at the anastomosis and progressing proximally will be recorded. To ensure complete small bowel scan the investigators recommend following parallel lines along the abdomen (extending from the liver/stomach/ribs to the bladder/iliac vessels).

Magnetic Resonance Enterography:

All MREs will occur at approved centers to ensure quality, using oral negative contrast and requiring previous fasting. The Baseline MRE with T1 and T2-weighted sequences will use intravenous contrast (gadolinium). MRE will be centrally read by established blinded experts, using the validated Lémann index score grading bowel damage for each segment.

Colonoscopy:

All participants with colonic involvement will undergo ileocolonoscopy (IC) after appropriate bowel preparation. IC will be locally read, using the validated definitions for the calculation of the Lémann index (Appendix 6) for each visualized segment. Where any segment of the IC is not directly visualized, reasons for limitations (e.g., impassable stricture, patient factors such as pain, etc.) will be recorded.

Cineloops of the entire withdrawal procedure should be stored locally, so central reading is still possible in case of any doubt.

Statistical Analysis:

Intraclass correlation (point estimate and 95% confidence interval) will be estimated using a two-way random model(6) for global, small bowel, colon, LI-IUS and LI-MRE, for global, colon LI-MRE + colonoscopy, for IUS and MRE segmental damage (terminal ileum and colonic segment, except rectum). These correlation coefficient estimates will be compared between IUS and MRE through maximum likelihood ratio-test (11).

IUS central review will be conducted by IBUS, an organization with vast experience and an established platform for ultrasound central review. There will be 3 MRE and 3 IUS central reviewers for this project.

All 102 cases will be evaluated by 3 central reviewers for inter-rater variability. All 102 cases will be evaluated twice by one of the central reviewers to determine intra-rater variability.

Intraclass correlation will be applied to estimate intra- and inter-reader reproducibility levels of global, small bowel, colon, LI-IUS and LI-MRE performances. When comparing global, small bowel, colon, LIIUS evaluations between local reader and central readers, the central readers data will be summarized by using either the common evaluation when at least 2 out of the 3 central readers agree, or the intermediate evaluation when the 3 readers disagree, omitting the most and the less severe evaluations.

For the other endpoints, standard nonparametric statistical methods will be used.