Introduction

Barrier healing (BH) has emerged as a novel treatment target in Inflammatory Bowel disease (IBD). Persistence of barrier impairment despite endoscopic and histological remission has been associated with worse long-term outcomes. However, accurate real-time assessment of intestinal barriers remains challenging in clinical practice. This study explores the role of advanced endoscopy (ultra-high magnification endocytoscopy [ECS] and confocal laser endomicroscopy [CLE]) for real-time, in vivo, structural and functional assessment of epithelial and vascular intestinal barrier. Further, we aim to investigate their correlation with automated barrier protein expression and clinical outcomes in IBD.

Aims & Methods

IBD patients undergoing endoscopic evaluation and healthy controls (HC) were recruited from two tertiary university centres.
The intestinal mucosa was assessed with ECS or CLE by novel scoring systems including epithelial and vascular features. BH was defined as ECS score ≤1 or CLE score ≤4, with preserved villi/crypt architecture, normal vessels, and, for CLE, absence of fluorescein leakage.
Biopsies were collected from both inflamed and non-inflamed areas. Epithelial (ZO-1, Claudin-2, E-cadherin) and vascular (PV-1 and CD-31) barrier markers were analysed through confocal microscopy, and their mean expression was automatically quantified using QuPath (v0.5.1). A biopsy-wise analysis assessed the correlation between endoscopic and histopathological evaluation. Moreover, occurrence of clinical outcomes at 6-month follow-up according to imaging-detected barrier impairment was evaluated.

Results

A total of 74 IBD patients (39 Crohn’s disease [CD], 35 Ulcerative colitis [UC]) and 9 HC) were considered. 46% (34/74) and 42% (30/74) IBD patients were in endoscopic and histological remission, respectively, while advanced imaging-defined BH was present in 24.3% (18/74) patients (Table 1). Barrier impairment was significantly associated with clinical outcomes at 6-month follow-up (p<0.01).
Mean expression of Claudin-2 was significantly higher in UC with ECS-based barrier impairment (85.5 vs 175.1, p 0.02), while PV1 expression was elevated in CD with CLE-based barrier dysfunction (362.1 vs 442.2, p 0.1). Significant correlations were found between ECS/CLE scores and epithelial and vascular barrier proteins.
In UC, Claudin-2 showed a moderate correlation with altered crypt architecture, goblet cell depletion and ECS total score (r 0.5), while PV-1 was moderately associated with altered vascular pattern and ECS total score (r -0.4).
In CD, Claudin-2 demonstrated a moderate correlation with crypt distortion, goblet cell depletion and total CLE scores (r 0.5), while PV-1 moderately correlated with fluorescein leakage, altered vessel architecture and total CLE score (r 0.4).

Conclusion

pCLE and ECS enable real-time intestinal barrier assessment in IBD, showing good correlation with epithelial and vascular barrier proteins and predicting clinical outcomes. This approach shows promise for targeting BH, guiding early and long-term disease management, and enabling flare-up prediction.

Introduction

Histological assessment in Inflammatory Bowel Disease (IBD) is often required with endoscopic diagnoses to establish disease activity. This is evident in patients where histological healing is a major therapeutic goal and when endoscopic healing can also exhibit microscopic disease. IBD can manifest in numerous landmark features which are structural and inflammatory related. Histological activity is determined through erosion/ulceration, architectural changes in tissue morphology, mucin depletion and presence of varied inflammatory cells the colonic crypts, Muscularis Mucosae (MM) and Lamina Propria (LP). Histological scoring systems vary in complexity and methodology and subject to discrepancies in inter-rater consensus. Our model aims to provide accurate measurements of key histological features in Ulcerative Colitis (UC) used in scoring systems as both a decision support tool for improving pathologist scoring alignment and sensitive quantitative metrics of inflammatory cells to evaluate disease activity.

Aims & Methods

We developed an ensemble of convolutional neural networks to identify cell types, tissue foreground, crypts, MM, and haemorrhage in H&E images of IBD biopsy data. Models were trained with expert-labelled annotations comprising 34,289 image-annotation pairs (80% training, 20% validation) from in-house and publicly available data. Model accuracy was assessed with 6,464 held-out image-annotation pairs from 26 Whole Slide Images (WSIs) from all sites of the large intestine commonly sampled during endoscopic investigation, while ground truth annotations were reviewed by GI-trained pathologists to certify accuracy. We used geospatial methods to derive boundaries of the LP from other predicted domains, associate cells to delineated regions, measure architectural changes and derive spatially differentiated metrics of inflammatory burden in 66 patients with confirmed UC.

Results

The AI model spatially differentiates and segments cell types and tissue compartments. This produces metrics for inflammatory cell counts in regions of interest such as crypts, LP, and MM. Goblet cell segmentation within crypts allows for measurement of mucin depletion. This can be visualized for pathologists where each inflammatory cell type across the WSI is represented as density metrics. Automated quantification of inflammatory cells across AI-identified tissue domains throughout the bowels of UC patients revealed large intra- and inter-patient variability.

Conclusion

Our AI model can measure histological healing by detecting architectural change, inflammatory cell number, goblet cell reduction and erosions and ulceration. This allows for metrics and visual overlays to assist with pathologist reading and quantifying changes in mucosal biopsies in a clinical trial setting which requires extremely sensitive tools only AI can provide.

Disclosure

DW/AM/RL/PA/PW and CL are employees at Perspectum Ltd. EF and RG are consultants for Perspectum Ltd. All other co-authors have no conflicts of interest to declare relevant to this work.

Introduction

The BIOSTOP Trial is a prospective, multicenter, randomized, controlled, open-label study conducted in Norway to evaluate the two-year outcomes of tumor necrosis factor-alpha (anti-TNF) antibody withdrawal in patients with ulcerative colitis (UC) in sustained clinical and endoscopic remission. Prior non-randomized studies have suggested that approximately half of such patients maintain remission after anti-TNF discontinuation and that reintroduction of anti-TNF therapy is generally effective and safe in the event of relapse.

Aims & Methods

A total of 174 UC patients in confirmed clinical and endoscopic remission were enrolled across 19 gastroenterology centers between June 2017 and December 2020 and randomized 1:1 to either continue or discontinue anti-TNF therapy. Eligibility criteria included at least 12 months of first-line anti-TNF maintenance treatment, clinical remission, fecal calprotectin <200 µg/g, and endoscopic remission (Mayo endoscopic score 0 or 1). Two-year follow-up data were available for 172 patients (86 in each group). In 156 of the patients endoscopies were available for central reading (77 in the withdrawal group and 79 in the maintenance group). The primary endpoint was sustained endoscopic remission at two years.

Results

At two years, central reading of follow-up colonoscopies confirmed endoscopic remission in 64 of 77 patients (83%, 95% CI 71.8 – 93.1) in the anti-TNF withdrawal group, compared to 77 of 79 patients (96 %, 95 % CI 92.1 – 100.8) in the anti-TNF maintenance group. The non-inferiority primary endpoint of the study was thus not met. During the study period, 48% of patients in the withdrawal group and 88% in the maintenance group remained in sustained clinical and endoscopic remission without relapse. Among the 45 patients in the withdrawal group who experienced endoscopic confirmed relapse, 28 (62%) resumed anti-TNF therapy, with 26 achieving both clinical and endoscopic remission. Additionally, 10 of 13 patients with mild relapses achieved endoscopic remission with optimized non-biologic therapy. There were no difference in serious adverse events between the groups.

Conclusion

After two years, 83% of patients in the anti-TNF withdrawal group and 96% in the maintenance group were in clinical and endoscopic remission, including those who experienced flares and achieved new remission after reintroduction of anti-TNF or optimized non-biologic therapy. Nearly half of the patients maintained remission without anti-TNF, and the majority of relapses were effectively managed. These findings suggest that anti-TNF withdrawal is a safe option for selected UC patients in confirmed remission, provided they are monitored closely.
Trial Registration: ClinicalTrials.gov Identifier: NCT03011268; EudraCT Number: 2016-001409-18. Trial completion date: 31 January 2025.

Disclosure

No disclosures