The CROCO (CROhn’s Disease COhort Study) – study design and protocol

Abstract

Background:

Crohn’s disease (CD) is a chronic, relapsing and remitting inflammatory bowel disease that can be associated with significant bowel damage and disability. The Lémann Index (LI) is a validated tool for measuring cumulative bowel damage in CD patients through a comprehensive assessment of stricturing, penetrating and surgical lesions. However, prospective studies evaluating bowel damage progression in recently diagnosed CD patients remain limited.

Objectives:

To characterise the absolute and longitudinal variations in bowel damage progression, as measured by the LI, in a cohort of recently diagnosed CD patients, and to assess its association with relevant disease features, including disease phenotype, treatment strategies, biomarkers and disability.

Design:

Study protocol for the Crohn’s Disease Cohort Study (CROCO Study), a multicentre, European, prospective cohort study.

Methods and analysis:

Patients with recently diagnosed CD (within the previous 12 months) will be enrolled and followed up for 5 years. Patients will receive standard-of-care treatment determined by the practising gastroenterologist. Morphological assessments to measure the LI and to evaluate bowel damage progression will be performed at years 1, 3 and 5 after the diagnosis. Disability will be assessed annually using the Inflammatory Bowel Disease – Disability Index (IBD-DI). The primary outcome will be the absolute LI at year 3 following diagnosis. Predictors of bowel damage progression and the association between bowel damage and disability will be analysed.

Discussion:

The CROCO study represents a unique multicentre cohort of recently diagnosed CD patients, designed to advance the understanding of CD’s natural history and evolution. It will facilitate the development of composite scores for predicting bowel damage progression and provide valuable tools for designing future disease-modification trials.

Trial registration:

NCT05420233.

Introduction

Crohn’s disease (CD) is a chronic inflammatory bowel disease (IBD) that follows a relapsing and remitting pattern, with varying phenotypes of disease progression over time (Figure 1). Despite advances in medical therapies and therapeutic strategies, a significant proportion of patients with moderate-to-severe CD experience disease progression, leading to high rates of surgery, hospitalisation and disability.^1–3 In addition, one-third of patients present with complications, such as stenotic and/or penetrating disease, at diagnosis. ² However, disease progression in CD is heterogeneous, and 20%–30% of patients may also have a mild disease course, with stable disease location, mild disease activity and no complications. ⁴

Figure 1.

Different phenotypes of disease progression and intestinal damage in Crohn’s disease.

CD also significantly impacts patients’ quality of life. As a result, disease modification, including regression of disability and reduction of the development of disease complications, such as bowel damage, is currently considered a key endpoint that should be incorporated into prospective disease-modification trials. ⁵

Many indices commonly used in clinical trials primarily reflect disease activity at a single time point, rather than the cumulative burden of the disease. The Lémann Index (LI) is a validated tool developed to measure the cumulative bowel damage in CD, which was recently updated by combining data from development and external validation studies involving 272 patients from 27 centres.^6,7 The LI assesses structural bowel damage comprehensively, including stricturing lesions, penetrating lesions (fistula and abscesses) and surgical resections. It is applicable in various settings, such as different CD locations and extent, early or advanced disease and patients with or without a history of surgery (Figure 2).^6,7 The LI has been proposed as an outcome in disease-modification trials.

Figure 2.

To calculate the Lémann Index, the entire gastrointestinal tract is divided into four organs (upper tract, small bowel, colon/rectum and anus). Each organ is then divided into different segments, and each segment is scored, using a 4-degree scale (0–3), according to the severity of stricturing and/or penetrating lesions and surgical interventions.

While the LI offers an objective and reproducible measure of cumulative bowel damage in CD, there remains a significant need for prospective assessment of the LI in a cohort of recently diagnosed CD patients to identify predictors of bowel damage progression that could guide therapeutic decisions. Currently, known predictors of poor outcomes in CD, identified from retrospective studies, include ileal and/or ileocolonic disease location, extensive small bowel disease, severe upper GI involvement, rectal disease, perianal lesions, early stricturing/penetrating complications, young age at diagnosis and smoking. ⁸ Smoking is also the most important risk factor for postoperative recurrence. ⁹ However, because most of the studies used varied definitions of severe disease, there is a need to validate these predictors of bowel damage progression.

Furthermore, there is limited data on the longitudinal characterisation of disability in early CD patients, which would also be valuable for future disease-modification trials.

To address these gaps, we have initiated the CROCO Study (Crohn’s Disease Cohort Study) to improve understanding of the long-term evolution of CD, assess the impact of different therapeutic strategies, and develop reliable predictors of bowel damage progression and disability. This protocol outlines our methods for prospectively following a cohort of newly diagnosed CD patients, with close monitoring of disease progression using the LI over time.

Study approach and design

Study objectives

The CROCO study is a prospective cohort study including recently diagnosed CD patients across multiple European centres. The overall aim of the CROCO study is to understand and better characterise bowel damage progression and to assess bowel damage progression in years 1 (Y1), 3 (Y3) and 5 (Y5) following CD diagnosis. The primary objective is to evaluate the absolute LI at Y3. As secondary objectives, the CROCO study will assess the absolute LI and its longitudinal variation throughout the study follow-up, identify predictors of bowel damage progression over time, examine the impact of different treatment regimens on the development of bowel damage, analyse the sensitivity of LI changes in response to treatment and explore serological predictors of LI variation along with the clinical course of CD in recently diagnosed patients. It will also assess disability, measured by the Inflammatory Bowel Disease–Disability Index (IBD-DI), and its predictors, and the association between disability and bowel damage progression. Furthermore, the CROCO study aims to implement and standardise examination reports and bowel damage definitions that are currently used to calculate the LI.

Study population

Adult patients (>18 years old) with recently diagnosed CD (within the past 12 months before inclusion) are eligible to participate in the study, provided they understand the information provided and give written informed consent for their participation. Patients with an uncertain diagnosis of CD, severe underlying medical conditions that anticipate a life expectancy of less than 2 years, or those who decline to participate are excluded. Furthermore, patients with medical conditions (e.g. gadolinium allergy, a glomerular filtration rate <30 mL/min or claustrophobia) that hinder the completion of cross-sectional imaging during follow-up are also excluded.

Patient recruitment was planned over 3 years across 19 European centres in Portugal, Spain, United Kingdom, Romania, Malta, France, Denmark, Czech Republic, Cyprus and Belgium. Enrolment was scheduled to begin once local ethical and institutional approvals had been obtained. To ensure balanced representation, each centre was asked to enrol between 20 and 40 patients, with flexibility to adjust these targets as needed throughout the study. Due to the COVID-19 pandemic, the study’s launch was delayed, eventually initiating in 2021, with several centres opening recruitment in 2022 and 2023.

Study design

This is a multicentre, European, prospective cohort study involving patients with recently diagnosed CD. The STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guideline was consulted when preparing this study protocol manuscript (Supplemental Material).

Patients are recruited into CROCO within the first 12 months following their diagnosis. At the time of inclusion, data on clinical history, biochemical parameters and the IBD-DI are collected. Baseline imaging, including magnetic resonance enterography (MRE), is not mandatory and may be performed at the discretion of the treating physician. Patients are subsequently followed every 6 months for up to 5 years, at which point the study concludes (Figure 3). During follow-up, patients receive standard-of-care treatment tailored to clinical and anatomical disease severity, as determined by the treating gastroenterologist. Anatomical and phenotypic disease progression is assessed through a minimum of three comprehensive morphological evaluations of the gastrointestinal tract during the study period. These evaluations include MRE and clinical examination for all patients, with additional procedures – such as upper gastrointestinal endoscopy, colonoscopy and/or pelvic MRE – performed as clinically indicated based on disease location. These assessments are used to calculate the LI at Y1, Y3 and Y5 following the diagnosis, providing an objective measure of the cumulative bowel damage. Disability is assessed annually, from baseline (inclusion date) through Y5, using the IBD-DI.

Figure 3.

CROCO study design.

CROCO, Crohn’s Disease Cohort Study; Fcal, faecal calprotectin; IBD-DI, inflammatory bowel disease–disability index; Labs, laboratory tests; LI, Lémann index.

Data collection

Data is collected and managed using REDCap (Research Electronic Data Capture) at each site by the local investigators.^10,11 REDCap is a secure, web-based software platform designed to support data capture for research studies, providing (1) an intuitive interface for validated data capture; (2) audit trails for tracking data manipulation and export procedures; (3) automated export procedures for seamless data downloads to common statistical packages and (4) procedures for data integration and interoperability with external sources.

At the time of recruitment (baseline visit), demographic data, medical history and clinical characteristics of the disease are collected. This includes details such as symptom onset, date of diagnosis, smoking status and family history of IBD. CD is classified according to the Montreal classification. Information on treatments received from diagnosis to inclusion, as well as any prior hospitalisations, perianal procedures or abdominal surgeries, is recorded.

Patients are followed every 6 months for 5 years. During these visits, blood samples are collected as part of routine care, and clinical disease activity is assessed using the Harvey-Bradshaw Index. Any interval event occurring since the previous visit, such as hospitalisations, perianal interventions, abdominal surgeries, pregnancies, cancer diagnoses or death, is documented. Moreover, details regarding the anatomical location and extent of any intestinal resections, as well as changes in medical treatment, are recorded.

Disability is assessed at baseline and then annually using the IBD-DI, which was translated into the local language of each participating country. ¹² Bowel damage is evaluated using the LI, based on morphological assessments performed at Y1, Y3 and Y5 following diagnosis.

Morphological assessments

At Y1, Y3 and Y5, morphological assessments of the digestive tract are performed according to the LI protocol using clinical evaluation, magnetic resonance and endoscopic examinations. ⁶ Morphological evaluations conducted at any other time points are optional and included for disease progression evaluation when performed.

Before the project began, a comprehensive teaching session was delivered to all investigators to standardise the morphological assessments. In addition, a remote site initiation visit was conducted at each centre to ensure consistency in procedures and a full understanding of the study protocol.

Morphological assessments are performed collaboratively by one gastroenterologist and one radiologist at each centre. The LI is calculated according to the validated and updated protocol, which divides the digestive tract into four organs: upper tract, small bowel, colon/rectum and anus. ⁷ Subsequently, each organ is divided into segments. MRE is performed on all patients. Upper gastrointestinal endoscopy is performed in cases of known or suspected involvement of the upper tract; colonoscopy is conducted for patients with known or suspected colonic and/or rectal disease and pelvic MRE is performed in the presence of perianal involvement.

For each segment, data on previous surgical interventions are collected, and the presence of stricturing and/or penetrating lesions is assessed and graded according to their maximum severity on a scale from 0 to 3 (Figure 2 and Supplemental Table 1). Rounded coefficients, based on both lesion severity and the organ involved, are applied to generate segmental indices. These are then combined using organ-specific coefficients to calculate a global score for each organ. The sum of these organ-specific scores constitutes the overall LI, providing a comprehensive measure of cumulative structural bowel damage.

Biobanking sub-study

A sub-study is being conducted with CROCO patients who have consented to collect biological samples, alongside clinical data. The purpose of this sub-study is to prospectively collect bio-specimens that could help understand disease pathogenesis, molecular/genomic phenotypes, clinical outcomes, predictors of disease course and response to therapy.

Biological samples are collected during the first year of enrolment, ideally at diagnosis and before the initiation of any treatment. If this is not feasible, samples are still collected, regardless of any therapies the patient has received or is currently undergoing, provided they are obtained within the first year after diagnosis. The goal is to identify biomarkers that may predict bowel damage (as measured by the LI) and/or disease progression, including the development of new CD complications such as strictures, fistulas or the need for surgery.

The required samples comprise: (1) peripheral blood for DNA, RNA and serum collection; (2) tissue biopsies including two biopsies from the terminal ileum, rectum and the most inflamed segment of the colon (if there is no colonic inflammation, biopsies are collected from the right colon) and (3) stools for microbiome, virome and fungome analysis.

From the 19 centres participating in the main CROCO study, 11 have been enrolled in the biobanking sub-study.

Study outcomes

Primary outcome

The primary outcome of the study is the absolute LI at Y3.

Secondary outcomes

The following secondary outcomes will be assessed: (1) absolute LI at Y1 and Y5; (2) LI longitudinal variations at Y3 and Y5 relative to Y1 and LI longitudinal variations at Y5 relative to Y3; (3) absolute IBD-DI and IBD-DI annual variations from baseline until Y5; (4) abdominal and perianal surgeries (both time-to-event and frequency); (5) hospitalisations (both time-to-event and frequency); (6) progression in CD behaviour from B1 to B2 and/or B3 and from B2 to B3 disease (both time-to-event and frequency) and (7) longitudinal values of maximal semestrial C-reactive protein and faecal calprotectin.

Data analysis

Continuous variables will be analysed using an unpaired t-test or Wilcoxon rank-sum test as appropriate, according to data distribution. Categorical variables will be assessed using a proportion analysis. For the primary outcome, analyses looking at predictors of bowel damage will be conducted by dividing the cohort into two groups above and below a cut-off value of the LI (based on prior studies, a cut-off of 2 was defined), which will allow the classification of patients into low bowel damage and high bowel damage groups. ¹³ Clinical factors to be tested for significance as predictors of bowel damage will be demographic characteristics at inclusion (gender, age, family history of IBD, smoking status), disease location, disease behaviour, age at diagnosis, absolute LI at inclusion, duration of clinical remission (in semesters), serological markers and duration and dosage of exposure to various treatments between date of inclusion and date of assessment. The association between these factors and the primary outcome will be assessed using univariate and multivariate logistic regression analysis. A similar analysis will be performed to assess the association of the same predictors with disability using the IBD-DI as a secondary outcome. Survival analysis and logistic regression analysis will be used to assess the impact of demographics, baseline disease characteristics and treatment modalities on clinically relevant outcomes such as surgery, hospitalisation and progression in disease behaviour. We will also assess the correlation of inflammatory biomarkers with the LI and IBD-DI and the association between bowel damage progression as assessed by the LI and disability using the IBD-DI.

Sample size calculation

According to the work by Gilletta et al., ¹³ we assumed that 55% of the patients will experience bowel damage defined as a LI above 2, at 3 years following diagnosis. A predictor of bowel damage occurrence can separate the sample into two groups of patients: those in proportion π who experienced bowel damage above 2 (high risk) at 3–4 years after inclusion and those in proportion 1−π whose LI remains below 2 (low risk) after the same follow-up. The association between the predictor and bowel damage occurrence can be characterised by a difference (Δ) between proportions of patients who experienced a LI over 2 in the high and low-risk groups at 3–4 years after inclusion. If we assume that π is 20% and that Δ is at least 20%, when using a two-sided test with a type I error of 5% to detect an association between bowel damage occurrence and a predictor with a power of 90%, the number of patients to be included is 407, leading to 59% and 39% of patients with a LI over 2 at 3–4 years in the high and low-risk groups, respectively, and to an odds ratio of 2.25. ¹⁴ If π varies between 20% and 80%, the sample size is 407 or less, and the odds ratio varies between 2.25 and 2.35. Original sample size calculation for the study was based on 30% loss to follow-up provided by the MICISTA registry (J. Cosnes’s, personal communication), which would result in 600 patients needed to be recruited. If we assume a 17.9% estimation of loss to follow-up at Y3, based on the current cohort recruitment as of September 2, 2024, 496 patients need to be recruited.

Significance and innovation

The therapeutic options for IBD are expected to expand rapidly in the coming years, with the ability to target new disease pathways. While advancing our understanding of biological mechanisms is essential to guide therapy selection, it will be equally important to stratify patients by disease severity and risk of progression. This reinforces the need to identify reliable predictors of disease progression.

Accumulating data suggests that early therapy in CD is crucial to improving disease outcomes.^15–17 However, not all CD patients experience the same rate of progression, with some exhibiting mild disease and remaining stable. Thus, there is currently a need for standardised outcomes to better evaluate bowel damage progression.

The CROCO registry is a prospective study aimed at characterising the natural history of CD from diagnosis. It seeks to assess various disease progression patterns and their predictors, using a validated index for bowel damage progression rather than indirect markers of disease progression, such as surgery, hospitalisation or therapy escalation. CROCO is the first study to assess the role of the LI longitudinally in a cohort of newly diagnosed CD patients.

Previous retrospective studies have assessed the sensitivity of the LI to change with bowel damage progression. For example, Gilletta et al. found in a cohort of 221 CD patients that over 50% had significant bowel damage (LI > 2) within 73 months of diagnosis. Notably, the first LI assessment was the only early factor predicting later damage, while the duration of clinically active disease and prior intestinal resection were associated with bowel damage progression. ¹³ Another cohort of 30 subjects who achieved clinical remission with anti-TNF therapy and were followed for a median of 32.5 months showed that an increase in the LI > 0.3 was associated with bowel damage progression, which predicted the need for major surgery in the long term (hazard ratio 0.19, p = 0.005). ¹⁸

However, most of these studies were retrospective and unicentre, making CROCO the first multicentre research project prospectively assessing LI variations in early CD patients and identifying predictors of cumulative bowel damage progression and therapy response. Additionally, the study will provide a longitudinal assessment of disability using a validated index, allowing for analysis of its predictors. This focus on disease-modification outcomes will provide valuable data for future clinical trials.

Moreover, recent studies suggest that early surgical intervention, even when associated with bowel damage, may be effective in halting disease progression over 5 years of follow-up. However, these studies lacked access to comprehensive follow-up data, such as endoscopy or MRE.^19,20 Our cohort will include a diverse group of CD patients from various geographic regions in Europe, presenting a unique opportunity to explore disease progression in CD. With detailed prospective clinical data collected over time, we aim to develop composite scores that integrate clinical, genetic and serological information to predict outcomes such as disease progression and disability. Improved prediction tools will inform disease-modification trials and enhance clinical care. Furthermore, the study will establish a research biobank, which will be a valuable resource for current and future research. The study results will be disseminated through publications in international peer-reviewed journals and presentations at national and international conferences.

Discussion and summary

Despite recent advancements in CD treatment, which emphasise the importance of early diagnosis and early biological treatment, alongside tight monitoring and treat-to-target strategies, bowel damage progression is still frequent, with a significant number of patients experiencing disease progression. Besides, current treatment goals in CD aim for long-lasting remission, with a normalised quality of life and disability. However, until recently, there were no measurement tools to objectively assess bowel damage progression and disability, which has hindered the accurate characterisation of these outcomes in CD. Moreover, there is sparse data on the longitudinal characterisation of these outcomes and, particularly, on how they can be modulated by significant disease phenotypes and treatment modalities. Also, few studies have prospectively followed cohorts of patients with recently diagnosed CD. Although not a population-based inception cohort study, the CROCO study is a prospective study designed to follow a large cohort of early CD patients from multiple referral and non-referral centres across Europe for 5 years, with the aim of providing insights into longitudinal bowel damage progression and identifying significant predictors of disease progression.