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.

Introduction

Obefazimod (Obe) is an oral, once-daily (QD), small molecule which enhances expression of microRNA-124 and has been studied in two Phase 2 induction trials and subsequent open-label maintenance studies [1-3] in patients (pts) with moderately to severely active ulcerative colitis (UC). Here we report efficacy and safety of two Phase 3, 8-week, induction trials in adult pts with UC from ABTECT-1 [NCT05507203] and ABTECT-2 [NCT05507216].

Aims & Methods

The multicenter, randomized, double-blind, placebo-controlled ABTECT trials enrolled pts with moderate-to-severe UC (defined as modified Mayo score (MMS)≥ 5, with rectal bleeding sub-score (RBS) ≥ 1 and centrally read endoscopic score >2) who had inadequate response, loss of response, or intolerance to at least one prior therapy (with no upper limit), including corticosteroids, immunosuppressants, biologics, S1P receptor modulators and/or JAK inhibitors. Pts were randomized 2:1:1 to Obe 50 mg QD (Obe-50), Obe 25 mg QD (Obe-25) or placebo (PBO) for 8 weeks. The primary endpoint was clinical remission (per MMS) and secondary endpoints included clinical response, endoscopic improvement, symptomatic remission, and histo-endoscopic mucosal improvement (HEMI).

Results

1272 pts were randomized and treated in ABTECT-1 (636) and ABTECT-2 (636). In both trials, baseline demographics and disease characteristics were similar between groups; 45.3% and 49.3% of pts had inadequate response to 1 or more advanced therapies. A significantly higher proportion of pts receiving Obe-50 (ABTECT-1:21.7%, ABTECT-2: 19.8%) versus PBO (2.5% and 6.3%) achieved clinical remission (Obe-50-PBO difference: ABTECT-1: 19.3%, p<0.0001; ABTECT-2: 13.4%, p=0.0001) and met all key secondary endpoints in both trials. A significantly higher proportion of pts receiving Obe-25 versus PBO achieved clinical remission (Obe-25-PBO difference: 21.4%, p<0.0001) and met all key secondary endpoints in ABTECT-1. In a pooled analysis, both Obe-50 and Obe-25 met all primary and secondary endpoints with nominal significance (p<0.0001) (Table). The overall rate of serious adverse events and treatment emergent adverse events (TEAEs) leading to study drug discontinuation for pts treated with Obe were similar to PBO. Proportions of pts who reported at least one TEAE in ABTECT-1 were 59.4%, 46.9%, and 53.2% for Obe-50, Obe-25, and PBO, respectively. For ABTECT2, TEAEs occurred in 61.0%, 50.9%, and 48.4% for Obe-50, Obe-25, and PBO, respectively. The most frequent TEAE was headache (Obe-50: 20.8-25.8%; Obe-25: 14.5-15.6%; PBO: 5.7). The headaches were mild, transient, short in duration and not a barrier to treat as evidenced by a low discontinuation rate of 0-1.6%. No signal was observed for serious, severe, or opportunistic infections or malignancies.

Table: 8-week induction efficacy of obefazimod in ABTECT-1 and ABTECT-2 Phase 3 studies, and pooled ABTECT-1 and ABTECT-2^‡

Conclusion

In both ABTECT induction trials, primary and secondary endpoints were met; obe treatment led to statistically significant improvements in clinical, endoscopic, symptomatic and combined endoscopic-histologic endpoints at week 8. Obe was well tolerated with no new safety signals identified.

References

1. Vermeire S, et al. J Crohns Colitis. 2023; 17: 1689-1697
2. Vermeire S, et al. The Lancet Gastroenterology & Hepatology. 2022; 7: 1024-1035.
3. Vermeire S, et al. J Crohns Colitis. 2025 ; 19 : jjaf074

Disclosure

Authors reports following potential conflicts of interest:
BES: Abbvie, Abivax, Adiso Therapeutics, Agomab, Alimentiv, Amgen, AnaptysBio, Arena Pharmaceuticals, Artugen Therapeutics, Astra Zeneca, Biolojic Design, Biora Therapeutics, Boehringer Ingelheim, Bristol Myers Squibb, Celltrion, Equillium, Enthera, Enveda Biosciences, Evommune, Ferring, Fzata, Galapagos, Genentech (Roche), Gilead Sciences, GlaxoSmithKline, Gossamer Bio, Imhotex, Index Pharmaceuticals, Innovation Pharmaceuticals, Kaleido, Kallyope, Lilly, Merck & Co., Microba, Microbiotica, Mitsubishi Tanabe, Mobius Care, Morphic Therapeutics, MRM Health, Nexus Therapeutics, Nimbus Discovery, Odyssey Therapeutics, OSE Immunotherapeutics, Janssen, Palisade Bio, Pfizer, Protagonist Therapeutics, Q32 Bio, Rasayana Therapeutics, Recludix Therapeutics, Reistone Biopharma, Sanofi, Sorriso Therapeutics, Spyre Therapeutics, Takeda, Target RWE,Teva, TLL Pharmaceutical, Tr1X, Trex Bio, Union Therapeutics, Ventyx Biosciences.
SV: AbbVie, Abivax, AbolerIS Pharma, AgomAb, Alimentiv, Arena Pharmaceuticals, AstraZeneca, Avaxia, BMS, Boehringer Ingelheim, Celgene, CVasThera, Cytoki Pharma, Dr Falk Pharma, Ferring, Galapagos, Genentech-Roche, Gilead, GSK, Hospira, Imidomics, Janssen, J&J, Lilly, Materia Prima, MiroBio, Morphic, MrMHealth, Mundipharma, MSD, Pfizer, Prodigest, Progenity, Prometheus, Robarts Clinical Trials, Second Genome, Shire, Surrozen, Takeda, Theravance, Tillots Pharma AG, Zealand Pharma.
LPB: Abbvie, Abivax, Adacyte, Alimentiv, Amgen, Applied Molecular Transport, Arena, Banook, Biogen, BMS, Celltrion, Connect Biopharm, Cytoki Pharma, Enthera, Ferring, Fresenius Kabi, Galapagos, Genentech, Gilead, Gossamer Bio, GSK, IAC Image Analysis, Index Pharmaceuticals, Inotrem, Janssen, Lilly, Medac, Mopac, Morphic, MSD, Nordic Pharma, Novartis, Oncodesign Precision Medicine, ONO Pharma, OSE Immunotherapeuthics, Pandion Therapeuthics, Par' Immune, Pfizer, Prometheus, Protagonist, Roche, Samsung, Sandoz, Sanofi, Satisfay, Takeda, Telavant, Theravance, Thermo Fischer, Tigenix, Tillots, Viatris, Vectivbio, Ventyx, Ysopia.
SD: AbbVie, Ferring, Hospira, Johnson & Johnson, Merck, MSD, Takeda, Mundipharma, Pfizer Inc, Tigenix, UCB Pharma, Vifor, Biogen, Celgene, Allergan, Celltrion, Sandoz, Boehringer Ingelheim.
PSD: Abbvie, Abivax, Adiso, Alimentiv, Bristol Meyer Squibb, Celltrion, Genentech, Geneoscopy, Janssen, Pfizer, Takeda.
MD: Abbvie, Abivax, Arena Pharmaceuticals, Astra Zeneca, Boehringer Ingelheim International GmbH, Bristol-Meyer Squibb, Eli Lilly and Company, F. Hoffmann-La Roche Ltd, Genentech Inc, Gilead, Janssen Pharmaceuticals, Merck, Pfizer Inc, Prometheus Biosciences, Takeda Pharmaceuticals.
HT: Abbvie, Abivax, Dr Falk Pharma, ferring, Galapagos, Microbiotica, MSD, Pfizer, Takeda.
BS: AbbVie, Abivax, Boehringer Ingelheim, Bristol Myers Squibb, Dr. Falk Pharma, Eli Lilly, Endpoint Health, Falk, Galapagos, Gilead, Janssen, Landos, Materia Prima, PredictImmune, Pfizer, and Takeda; AlfaSigma, CED Service GmbH, MSD, Ferring, Galapagos, Tr1x bio
TH: Mitsubishi Tanabe Pharma Corporation, EA pharma Co. Ltd., AbbVie GK, JIMRO Co. Ltd., Zeria Pharmaceutical Co. Ltd., Kyorin Pharmaceutical Co. Ltd., Nippon Kayaku Co. Ltd., Takeda Pharmaceutical Co. Ltd., Pfizer Inc., Mochida Pharmaceutical Co. Ltd., Boston Scientific Corporation, Kissei Pharmaceutical Co. Ltd, Janssen Pharmaceutical K.K., Pfizer Inc., Eli Lilly, Gilead Sciences, Bristol Myers Squibb, Abivax.
XT: Celltrion, AbbVie, Johnson & Johnson, Lilly, Takeda, Alpha Sigma, Dr. Falk, Abivax, Biogen, Fresenius Kabi, MSD, Pfizer, Tillotts, and Thabor Therapeutics.
RA: AbbVie, Abivax, AstraZeneca, Bristol-Myers Squibb, Celltrion Healthcare, Galapagos, Johnson&Johnson, Lilly, MSD, Pfizer, and Takeda Pharma
US: AbbVie, Abivax, Amgen, Galapagos, Janssen, Eli Lilly, Boehringer Ingelheim, Bristol Myers Squibb, Celgene, Gilead Sciences, Pfizer, Roche,Takeda Pharmaceuticals
AA: AbbVie, Abivax, AG Pharma, Alfa Sigma, Astra Zeneca, Biogen, Boehringer Ingelheim, Bristol-Myers Squibb, Celltrion, Eli-Lilly, Enthera, Ferring, Galapagos, Gilead, Giuliani, Janssen, Lionhealth, MSD, Nestlé, Novartis, Pfizer, Protagonist Therapeutics, Roche, Samsung Bioepis, Sanofi, Sandoz, Takeda, Teva Pharmaceuticals, Tillots Pharma
FB: AbbVie, Amgen, Eurogenerics, J&J, Arena Pharmaceuticals, Celltrion, Ferring, Galapagos, Janssen, Merck Sharp & Dohme, Pfizer Inc, Takeda, Celgene, Fresenius Kabi, Sandoz
DTR: Abbvie, Abivax SA, Altrubio, Athos Therapeutics, Inc, Bristol-Myers Squibb, Celltrion, Connect BioPharma, Eli Lilly & Co., Genentech (Roche) Inc., Iterative Health, Janssen Pharmaceuticals, Johnson & Johnson, Merck & Co., Odyssey Therapeutics, Pfizer, Sanofi, Spyre, Takeda Pharmaceuticals, Vedanta Biosciences, and Ventyx.

Introduction

Pregnancy is common in women with inflammatory bowel disease (IBD), a population in whom optimal obstetric management is essential to ensure maternal and neonatal well-being. Vaginal delivery is generally recommended as the preferred mode of birth due to its lower risk of complications and faster recovery. Unnecessary caesarean section (C-section) delivery increases the risk of maternal complications, such as infection and thrombosis, and may negatively impact neonatal outcomes, including respiratory distress and delayed breastfeeding. Identifying the precise prevalence of C-sections and the factors driving that prevalence in women with IBD is essential to improve care and promote safe, evidence-based delivery planning.

Aims & Methods

To assess the frequency of C-section, and to identify factors associated with the mode of delivery in pregnant women with IBD included in the DUMBO registry.
Patients from DUMBO registry, a prospective multicentre cohort study of pregnant women with IBD in Spain, were included. The registry systematically collects demographic characteristics, obstetric history, IBD phenotype and activity, treatments received during pregnancy, and pregnancy-related complications. Mode of delivery is recorded for all participants. In cases of C-section, investigators document whether the indication was obstetric or IBD-related, as well as the specific reason for the procedure. Logistic regression analysis was performed to identify clinical and disease-related variables independently associated with the likelihood of C-section.

Results

A total of 1,012 women had given birth at the time of data analysis; mean age 34 years (standard deviation 4.8); median IBD duration 8.8 years (interquartile range 4.5–13); 626 (57%) had Crohn’s disease (CD) (21% had perianal disease, 19% active), 216 (19%) had previous surgery due to IBD, 464 (42%) were exposed to biologics, and 122 (11%) had active disease during pregnancy. A total of 293 deliveries (29%) were by C-section. Among C-sections, 237 (81%) were indicated for obstetric reasons, and 56 (19%) for IBD. In univariate analysis, older maternal age, longer IBD duration, to have CD (vs. ulcerative colitis), prior IBD surgery, active perianal disease, and exposure to biologic therapy during pregnancy were significantly more frequent among patients undergoing C-section compared to those with vaginal delivery (table 1). In the multivariate analysis, older maternal age [Odds ratio (OR) = 1.05, 95% confidence interval (CI) = 1.02–1.08], biologic treatment during pregnancy (OR = 1.46, 95%CI 1.08–1.95), prior IBD surgery (OR = 1.47, 95%CI 1.47–2.95), and active perianal disease (OR = 4.89, 95% CI 1.86–12.8) were independently associated with a higher risk of C-section.

Conclusion

One-third of deliveries in women with IBD are by C-section (a figure higher than that in general population¹). In 20% of cases, the indication for C-section was directly related to IBD-specific conditions. Several factors associated with an increased risk of C-section were identified, which are also recognized as predictors of more severe disease. Further efforts are needed to assess the appropriateness of C-section indications and to reduce unnecessary procedures in women with IBD.

References

¹ Ministerio de Sanidad. Annual report on the National Health System of Spain 2023 [Internet]. Madrid: Ministerio de Sanidad; 2023 [accessed 2025 April 14]. p. 86. Available from: https://www.sanidad.gob.es/estadEstudios/estadisticas/sisInfSanSNS/tablasEstadisticas/InfAnualSNS2023/INFORME_ANUAL_2023.pdf

Disclosure

María José Casanova María José Casanova, has received education or research funding from Pfizer, Takeda, Janssen, MSD, Dr. Falk, Shire, Ferring, Galápagos, Abbvie and Lilly.
María Chaparro has served as speaker, consultant or research or education funding from MSD, Abbvie, Hospira, Pfizer, Takeda, Janssen, Ferring, Shire Pharmaceuticals, Dr. Falk Pharma, Tillotts Pharma, Biogen, Gilead and Lilly.
Javier P. Gisbert has served as speaker, consultant, and advisory member for or has received research funding from MSD, Abbvie, Pfizer, Kern Pharma, Biogen, Mylan, Takeda, Janssen, Roche, Sandoz, Celgene/Bristol Myers, Gilead/Galapagos, Lilly, Ferring, Faes Farma, Shire Pharmaceuticals, Dr. Falk Pharma, Tillotts Pharma, Chiesi, Casen Fleet, Gebro Pharma, Otsuka Pharmaceutical, Norgine and Vifor Pharma.
Cristina Calvino-Suarez, has served as a speaker for or has received research funding from MSD, AbbVie, Janssen, Takeda and Adacyte.
Pilar López Serrano has received grants from MSD, Abbvie, Takeda, Janssen, Faes, Ferring and Dr Falk Pharma.
Noemí Manceñido Marcos has served as a speaker and consultant for or received financial support for educational activities from Janssen, Abbvie, Pfizer, Takeda, Ferring, Faes Farma, Dr. Falk Pharma, and Tillots Pharma.
Marta Maia Boscá Watts Conflictos de interés: MSD, Ferring, Abbvie, Janssen, Biogen, Pfizer, Galápagos, Otsuka, Kern pharma y Takeda.
Agnès Fernández-Clotet has served as a speaker, or has recieves education funding from Dr. Falk, Janssen, Takeda, Chiesi and Pfizer.
Lara Arias García has served as speaker or has received research or education funding from MSD; Abbvie; Kern Pharma; Ferring; Faes Farma; Shire Pharmaceuticals; Pfizer; Takeda; Janssen; Tillots Pharma and OtsuKa Pharmaceutical.
Lucía Marquez-Mosquera has been served as speaker and / or has received research or education funds from Abbvie, Pfizer, Takeda, Janssen.
Marta Calvo Moya has served as a speaker, consultant and advisory member for or has received research funding from MSD, Abbvie, Takeda, Jannsen, Pfizer, Lilly, Mylan, Kern Pharma, Ferring, Shire Pharmaceuticals, Tillots, Dr. Falk Pharma, Chiesi, and Gebro Pharma.
Miriam Mañosa has served as a speaker, consultant and advisory member for or has received research funding from AbbVie, Janssen, MSD, Pfizer, Faes, Takeda, Tillots, Ferring.
Vicent Hernández has served as consultant, has served as speaker, has received travel support or research funding from MSD, AbbVie, Ferring, Dr. Falk Pharma, Tillotts Pharma, Pfizer, Takeda, Janssen, KernPharma Biologics, Adacyte, Sandoz, Biogen, Fresenius-Kabi, FAES Farma and Galapagos.
Raquel Camargo-Camero, He recibido financiación para diferentes tipos de actividades formativas de Janssen, Ferring, Abbvie, Falk, Pfizer, Galápagos y Takeda.
Iván Guerra has served as speaker or has received education funding from Takeda and Tillots.
David Busquets declares educational activities, research projects and scientific meetings sponsored by Abbvie, Janssen, Ferring, Pfizer, Tillots Pharma and Takeda.
Montserrat Rivero has served as a speaker or advisory member from TaKeda, Janssen, Galapagos, Ferring and Pfizer.
Iago Rodríguez-Lago has received financial support for traveling and educational activities from or has served as an advisory board member for Abbvie, Adacyte, Celltrion, Chiesi, Danone, Ferring, Faes Farma, Janssen, Galapagos, MSD, Pfizer, Roche, Takeda, and Tillotts Pharma. Financial support for research: Tillotts Pharma.
Rest of authors have nothing to declare.

Introduction

JAK inhibitors (JAKi) are small molecules capable of crossing the placenta via passive diffusion or active extracellular transport from early pregnancy.¹ Data on the risks of in utero JAKi exposure are limited,² particularly regarding pregnancy outcomes, infant infections, vaccine responses, and childhood development.

Aims & Methods

We conducted a multicenter retrospective JAKi and pregnancy observational study supported by the ECCO Collaborative Network for Exceptionally Rare Case Reports (CONFER). Pregnant women with inflammatory bowel disease (IBD) exposed to tofacitinib (TOFA), upadacitinib (UPA), or filgotinib (FILGO) during pregnancy were enrolled from 21 centers across 15 countries. Data collected included maternal demographics, IBD characteristics, disease activity, medication use - including timing and duration of JAKi exposure - along with maternal, pregnancy, and infant outcomes.

Results

Of 43 JAKi exposed pregnancies, 17 (39.5%) were unplanned. Disease activity was present in 22 (51.2%) of the pregnancies. One-in-five women initiated JAKi in pregnancy (n=8, 18.6%). Abortion occurred in 8 cases, at a median gestational age (GA) of 8.5 weeks (range 3-16). Of these, 3 (7.0%) were spontaneous and 5 (11.6%) induced, with two terminations due to concerns about potential adverse effects of JAKi treatment. Of the remaining 35 pregnancies 26 (74.3%) were exposed to TOFA (5 mg BID (n=9), 10 mg BID (n=17)), 8 (22.9%) to UPA (30 mg daily (n=2), 45 mg daily (n=6)) and one (2.9%) to FILGO (200 mg daily). The majority received JAKi treatment throughout pregnancy (n=24, 68.6%). Discontinuation of JAKi during the 1^st or 2^nd trimester was followed by disease activity in 7 of 11 cases (63.6%). Maternal complication risk was low: no cases of DVT or (pre)eclampsia; premature rupture of membranes (GA 34–37) occurred in 3 cases, and 4 women required hospitalization due to infection or other complications. Among 35 live births, one infant (2.9%) was preterm (GA 34), three (8.6%) were small for gestational age, one (2.9%) had an Apgar score <7 at 5 minutes, and no congenital abnormalities were reported. Follow-up infant data are shown in table 1. One-third of breastfeeding women continued JAKi treatment. A high adherence rate to the national immunization schedule was seen. Three (8.6%) skipped the live rotavirus vaccine due to in uteroJAKi exposure. No adverse effects were reported following live attenuated or inactivated vaccines, and no malignancies were observed.

Conclusion

In this global multicenter study of JAKi exposure during pregnancy in women with IBD, maternal complication rates were low, and no increased risk of adverse pregnancy outcome were observed among live births. While disease activity was common, particularly following early discontinuation of JAKi, infant outcomes - including development, infection risk, and vaccine response - were reassuring. These findings provide important preliminary safety data to guide decision-making regarding the use of JAKi in pregnancy, though further prospective studies are needed to confirm long-term safety.

References

1. Fein KC, Arral ML, Kim JS, et al. Placental drug transport and fetal exposure during pregnancy is determined by drug molecular size, chemistry, and conformation. J Control Release 2023;361:29-39.
2. Torres J, Chaparro M, Julsgaard M, et al. European Crohn's and Colitis Guidelines on Sexuality, Fertility, Pregnancy, and Lactation. J Crohns Colitis 2023;17:1-27.

Disclosure

M. Julsgaard: Advisory board for AbbVie. G. Santana: Speaker and advisory board for pfizer.

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.