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

Portal hypertension (PHT) most commonly due to cirrhosis, but non-cirrhotic portal hypertension (NCPH) related to metabolic dysfunction, specifically metabolic dysfunction-associated steatotic liver disease (MASLD), necessitates further clarification of its mechanisms and identification of therapeutic targets.

PHT may precede the development of hepatic fibrosis in MASLD, and even serve as a driving factor in the pathogenesis of MASLD. Consequently, metabolic dysfunction-associated PHT might not depend on structural alterations such as hepatic fibrosis or pseudolobule formation, but rather correlate with other factors such as endothelial dysfunction, which has been demonstrated to play a pivotal role in NCPH. Endothelial dysfunction encompasses endothelial injury and endothelial-to-mesenchymal transition (EndMT), the latter being a specific type of epithelial-to-mesenchymal transition (EMT). As reported, the JAK/STAT3 signaling pathway plays a pivotal role in the initiation and progression of EMT. Similarly, patients with NCPH induced by myeloproliferative neoplasms (MPN) often lack pathological features such as hepatic fibrosis or pseudolobule formation. The Janus kinase 2 (JAK2) V617F mutation is highly prevalent in MPN patients. Therefore, targeted inhibition of JAK2 serves as an effective therapeutic strategy for MPN.

Given the shared pathological features between MPN-associated portal hypertension and MASLD-associated portal hypertension, as well as the vital​ role of JAK2 in both disease, the application of JAK2 inhibitor holds significant therapeutic potential for metabolic dysfunction-associated portal hypertension.

Aims & Methods

This study aims to verify the therapeutic efficacy of JAK2 inhibition in attenuating MASLD-associated PHT, with particular focus on its modulation of EndMT. Mice were fed with a high-fat diet (HFD) diet for 12 weeks to induce MASLD-associated PHT and divided into two groups, the treatment group (Rux) and control (Ctr) group. Ruxolitinib (JAK2 inhibitor) was given as a suspension by gavage (60 mg/kg in 0.2 ml) every two days for 3 weeks, while controls received the same volume of saline, from week 10-12 of feeding. Portal pressure was measured, serological analysis and histopathological staining were performed. RT-PCR and Western Blot were conducted to measure EndMT-related genes expression.

Results

The treatment group (Rux) demonstrated a significant reduction in directly measured portal pressure, with concordant changes observed in the spleen-to-body weight ratio, which served as an indirect indicator of portal pressure. Concurrent improvement was noted in ALT, AST and TC. H&E, Masson and Sirius red staining showed decreased perisinusoidal and perivascular fibrosis in the Rux group. Electron microscopy revealed restoration of endothelial fenestration. Additionally, ruxolitinib treatment led to downregulation of endothelial damage markers, mesenchymal cell markers as well as EndMT-related regulatory factors, indicating an alleviation of EndMT and endothelial dysfunction.

Conclusion

The JAK2 inhibitor demonstrates therapeutic potential for MASLD-associated PHT through modulation of endothelial dysfunction and EndMT. These mechanistic insights position ruxolitinib as a promising candidate for clinical management of NCPH, particularly in the MASLD-associated PHT.

References

[1] European Association for the Study of the Liver; European Association for the Study of Diabetes; European Association for the Study of Obesity. EASL-EASD-EASO Clinical Practice Guidelines on the management of metabolic dysfunction-associated steatotic liver disease (MASLD): Executive Summary. Diabetologia. 2024 Jun 13. [2] Song Z, Liu X, Zhang W, et al. Ruxolitinib suppresses liver fibrosis progression and accelerates fibrosis reversal via selectively targeting Janus kinase 1/2. Journal of Translational Medicine. 2022, 20(1): 157.
[3] Zhang XL, Zhang XY, Ge XQ, et al. Mangiferin prevents hepatocyte epithelial-mesenchymal transition in liver fibrosis via targeting HSP27-mediated JAK2/STAT3 and TGF-β1/Smad pathway. Phytotherapy Research. 2022, 36(11): 4167-4182.
[4] Jin W. Role of JAK/STAT3 Signaling in the Regulation of Metastasis, the Transition of Cancer Stem Cells, and Chemoresistance of Cancer by Epithelial-Mesenchymal Transition. Cells. 2020;9(1):217.
[5] Hammoutene A, Rautou PE. Role of liver sinusoidal endothelial cells in non-alcoholic fatty liver disease. Journal of Hepatology. 2019, 70(6): 1278–1291.

Introduction

The achievement of histological remission (HR) is a key treatment goal in ulcerative colitis (UC), as it is suggestive of better disease management and predicts response to therapy.
The absence of neutrophils in epithelium and lamina is crucial to define HR. However, the role of neutrophil number and localisation remains unclear. We aimed to develop a novel artificial intelligence (AI)-driven system to automate neutrophil detection, localisation, and quantification, supporting assessment of HR and enable the prediction of response to therapy.

Aims & Methods

We developed an AI-driven system by integrating two deep learning models to (1) segment whole slide images (WSIs) into epithelium, crypts and lamina propria and to (2) detect and quantify neutrophils. Outputs from these models were combined to compute neutrophil densities in the lamina propria, epithelium, and overall tissue. The optimal neutrophil density cut-offs to assess disease activity (Geboes score >2B.0) in UC patients were identified in the AMAC phase 2 Mirikizumab trial, and subsequently validated in the multicentre prospective PICaSSO cohort study. Cut-offs to determine treatment response at weeks 12 and 52 (histological improvement [HI; Geboes score ≤3.1] and histological remission [HR; Geboes score ≤2B.0]), were also evaluated.

Results

Table 1 shows the system diagnostic performance in assessing disease activity and predicting response to therapy at W12 and W52 in the Mirikizumab cohort.
303 WSIs from active UC patients from the multicentre, randomised, double-blind, parallel-arm, placebo-controlled phase 2 clinical trial of Mirikizumab were analysed. The model yielded a DICE SØrensen of 0.57 for segmentation, precision of 82.3% and recall of 77.0% for neutrophil detection. The system identified an overall optimal neutrophil density cut-off (cells/mm²) predictive of disease activity of >21.7.0 (>10.1 and >30.4 for epithelium and lamina propria, respectively), with an accuracy of 86% (85% and 85%, respectively). Further validation in the multicenter PICaSSO cohort yielded to similar results.
Moreover, the overall optimal cut-off to predict histological remission at W52 was <20.8 (<10.4 and <27.9 for epithelium and lamina propria, respectively), with an accuracy of 82.0% (81% and 80%, respectively).

Conclusion

Our novel AI-based system demonstrates strong potential as an automated tool to detect, localise, and quantify neutrophils to assess histological activity and predict response to therapy in a clinical trial UC cohort and real-world settings. This framework could be applied in clinical trials and routine clinical practice, offering objective guidance for personalised management in UC patients.

References

[1] Santacroce G, Meseguer P, Zammarchi I, et al. P406 A novel active learning-based digital pathology protocol annotation for histologic assessment in Ulcerative Colitis using PICaSSO Histologic Remission Index (PHRI). Journal of Crohn's and Colitis, Volume 18, Issue Supplement_1, January 2024, Pages i843–i844
[2] Graham S, Vu QD, Jahanifar M, et al. CoNIC Challenge: Pushing the frontiers of nuclear detection, segmentation, classification and counting. Med Image Anal. 2024;92:103047.

Disclosure

MI received research grants and consultant fees from Olympus, Pentax, Eli Lilly, and Janssen. All other authors declare no competing interests.

Introduction

Up to 40% of patients with acute severe ulcerative colitis (ASUC) do not respond to infliximab (IFX)(1) presumably due to insufficient drug exposure (1, 2). We investigated whether personalised IFX induction using predefined serum concentrations was superior to standard dosing.

Aims & Methods

In this prospective, multi-centre trial, adult IFX-naive steroid-refractory ASUC patients were randomised 1:1 to standard (SD) or personalised IFX dosing (PD). After an initial 5 mg/kg IFX infusion, SD was 5 mg/kg IFX at week 2 and 6. In PD, additional 5 mg/kg IFX infusions were administered guided by a Bayesian pharmacokinetic algorithm (iDose™) aiming at IFX serum concentrations >28 ug/mL day 0-28 and >15 ug/mL day 29-42 (3). The primary composite endpoint was clinical and endoscopic response at day 42 (Lichtiger score <10 and ≥3 points drop and UCEIS ≥2 points drop from baseline). Video-endoscopies were performed at baseline, day 42 and 182 and scored by 2 expert readers (XR) with adjudication by a third XR and post hoc also by the DovaVision UC AI tool (AI-R)(4). Key secondary endpoints were day 42 clinical response, day 42 endoscopic response, day 182 clinical remission (Lichtiger ≤3), day 182 endoscopic remission (UCEIS ≤1 on all components), and safety.

Results

48 patients were randomised (23 PD/25 SD) and 31 completed the 26-week study (19 PD/12 SD). Patient characteristics were comparable. Median cumulative IFX dose until day 42 was 18.41 mg/kg [1.77, 20.27] for PD vs 13.79 mg/kg [10.38, 14.82] for SD. The primary composite endpoint at day 42 was not met with XR (57% in PD vs 44% in SD; p=0.564). However, AI-R showed a significantly higher composite response rate in PD (74% in PD vs 32% in SD; p=0.0047). PD showed higher day 42 clinical response vs SD (91% vs 64%; p=0.039). Day 42 endoscopic response was observed in 57% in PD vs 44% in SD (p=0.564) with XR and 74% in PD vs 32% in SD with AI-R (p=0.0047). The proportion of patients in clinical and endoscopic remission at day 182 was higher in PD than in SD (65% vs 36%; p=0.082). Day 182 endoscopic remission with AI-R was higher in PD (52% in PD and 20% in SD; p=0.0337). SAEs occurred in 9% of patients on PD vs 20% of patients on SD. Following an interim analysis, the trial was discontinued based on futility.

Table 1. Baseline characteristics

Conclusion

Personalised IFX dosing was superior to standard dosing in ASUC when endoscopies were read by AI. This is the first IBD study where the primary endpoint, missed by expert assessment, was overturned and met by AI assessment instead. This demonstrates the potential of AI to fundamentally improve clinical trial methodology.

References

1. Seow CH, Newman A, Irwin SP, Steinhart AH, Silverberg MS, Greenberg GR. Trough serum infliximab: a predictive factor of clinical outcome for infliximab treatment in acute ulcerative colitis. Gut. 2010;59(1):49-54.
2. Papamichael K, Van Stappen T, Vande Casteele N, Gils A, Billiet T, Tops S, et al. Infliximab Concentration Thresholds During Induction Therapy Are Associated With Short-term Mucosal Healing in Patients With Ulcerative Colitis. Clin Gastroenterol Hepatol. 2016;14(4):543-9.
3. Ungar B, Mazor Y, Weisshof R, Yanai H, Ron Y, Goren I, et al. Induction infliximab levels among patients with acute severe ulcerative colitis compared with patients with moderately severe ulcerative colitis. Aliment Pharmacol Ther. 2016;43(12):1293-9.
4. Byrne M, Requa J, Panaccione R, Panes J, Bressler B, East JE, et al. Development and Validation of a novel AI-based computer vision solution for Ulcerative Colitis severity scoring on video for real world using high-volume expert-annotated video frames. United European Gastroenterol J. 2025 Oct.

Disclosure

This project received grant support from Pfizer Inc. and research support from Baysient LLC, Buhlmann Laboratories and Dova Health Intelligence.