Introduction

Anti-tumor necrosis factor (TNF)-α are the first target therapies approved for the treatment of steroid-dependent or -refractory ulcerative colitis (UC). Drugs with different mechanisms of action have recently entered the market, but data on the optimal management of UC after failure of first-line anti-TNF-α are lacking.

Aims & Methods

The PODIUM study is a multicentre, retrospective, observational study that aims to compare the real-life effectiveness and safety of second-line vedolizumab (VDZ), ustekinumab (UST) and tofacitinib (TOF) in patients with UC. Consecutive patients with UC who had failed at least one anti-TNF-α and were never exposed to other target therapies were enrolled from 12 centres across Europe. Sociodemographic data, data on clinical activity and adverse events of interest (AEIs, defined as AE requiring temporary/permanent drug discontinuation or medical intervention) were retrospectively extracted from medical records, from the initiation of second-line therapy until one year of observation/drug discontinuation/last available follow-up. The primary outcome was the cumulative probability of achieving steroid-free clinical remission (SFCR), defined as stool frequency=0-1 and rectal bleeding=0 and no concomitant steroids, over the first year of treatment; inverse probability of treatment weighting (IPTW), calculated as the inverse of the probability of receiving each drug (IPTW=1/Propensity Score), was applied to the primary outcome to account for imbalances in cohorts’ characteristics. We performed an interim analysis including all data collected up to March 15^th 2024; data collection is still ongoing at the submission of this Abstract.

Results

Data on 303 patients (172 VDZ, 63 UST, 68 TOF) were included in the analysis. The main baseline characteristics are presented in Table 1. After IPTW, the cumulative probability of achieving SFCR during the first year was 51.0%, 59.1% and 64.1% for VDZ, UST and TOF, respectively; both UST and TOF were superior to VDZ at Log-rank test (p<0.001), with no differences between the two. At weighted logistic regression, female sex (adjusted Odds Ratio, aOR, 0.06, Confidence Interval, CI, 95% 0.41-0.73, p<0.001) and higher Charlosn Comorbidity Index calculated without considering age (aOR 0.73, 95% CI 95% 0.60-0.88, p=0.001) were independently associated with a lower probability of achieving SFCR during the first year. Recorded AEIs were 16 (4 infections) in the VDZ group (9.3% of VDZ-treated patients), 6 (1 infection) in the UST group (9.5% of UST-treated patients), and 15 (5 infections, including 1 Zoster reactivation) in the TOF group (22.1% of TOF-treated patients).
Table 1

Conclusion

This preliminary analysis suggests that both UST and TOF might have superior effectiveness, compared to VDZ, as second-line agents in patients with UC after anti-TNF-α. Confirmatory analysis with a larger sample size will be performed. A slightly higher rate of AEIs was observed in the TOF group.

Disclosure

Giuseppe Privitera received consultancy fees from Alphasigma and Janssen. Cristina Bezzio received lecture fees and served as a consultant for Takeda, MSD, Ferring, Abbvie, Galapagos and Janssen. Edoardo Vincenzo Savarino has served as speaker for Abbvie, Agave, AGPharma, Alfasigma, Aurora Pharma, CaDiGroup, Celltrion, Dr. Falk, EG Stada Group, Fenix Pharma, Fresenius Kabi, Galapagos, Janssen, JB Pharmaceuticals, Innovamedica/Adacyte, Malesci, MayolyBiohealth, Omega Pharma, Pfizer, Reckitt Benckiser, Sandoz, SILA, Sofar, Takeda, Tillots, and Unifarco; has served as a consultant for Abbvie, Agave, Alfasigma, Biogen, Bristol-Myers Squibb, Celltrion, DiademaFarmaceutici, Dr. Falk, Fenix Pharma, Fresenius Kabi, Janssen, JB Pharmaceuticals, Merck & Co, Nestlè, Reckitt Benckiser, Regeneron, Sanofi, SILA, Sofar, Synformulas GmbH, Takeda, and Unifarco; he has received research support from Pfizer, Reckitt Benckiser, SILA, Sofar, Unifarco, Zeta Farmaceutici. Daniela Pugliese received speaker fees and/or advisory board from AbbVie, Galapagos, MSD, Takeda and Janssen, Pfizer. Iago Rodriguez-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. Manuel Barreiro-de Acosta has served as a speaker, consultant and advisory member for and has received research funding from MSD, AbbVie, Janssen, Kern Pharma, Celltrion, Takeda, Gilead, Celgene, Pfizer, Sandoz, Biogen, Fresenius, Ferring, Faes Farma, Dr. Falk Pharma, Chiesi, Gebro Pharma, Adacyte and Vifor Pharma. Sara Onali has received speaker fees from AbbVie, Takeda, Janssen, Amgen, Galapagos and Norgine. Massimo Claudio Fantini has received speaker and consultancy fees from AbbVie, Takeda, Janssen-Cilag, Pfizer, Sandoz, Biogen, Galapagos, Bristol-Myers Squibb, and MSD; and research grants from Pfizer and Janssen-Cilag. Karmiris K has served as speaker, consultant, and advisory member for Abbvie, Amgen, Enorasis, Ferring, Galenica, Genesis, Janssen, MSD, Pfizer, Takeda and Vianex. Henit Yanai has received consulting fees Abbvie, Ferring, Janssen, Neopharm Ltd., Pfizer, Takeda; Payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events Abbvie, Janssen, Pfizer, Takeda, Neopharm Ltd.; Participation on a Data Safety Monitoring Board or Advisory Board Abbvie, Neopharm Ltd., Pfizer, Takeda. Ana Gutierrez has served as speaker, consultant and advisory member for or has received research funding from MSD, Abbvie, Pfizer, Kern Pharma, Takeda, Janssen, Ferring, Faes Farma, Shire Pharmaceuticals, Tillotts Pharma, Chiesi and Otsuka Pharmaceutical. Luca Pastorelli has received consultant and/or speaker fees from Giuliani, Ferring, Janssen, Takeda. María Chaparro has served as a speaker and consultant for and has received research or education funding from MSD, AbbVie, Hospira, Pfizer, Takeda, Janssen, Ferring, Shire Pharmaceuticals, Dr. Falk Pharma, Tillotts Pharma, Biogen, Gilead and Lilly. Yamile Zabana has received speaker and/or consultant fees from AbbVie, Adacyte, Almirall, Amgen, Dr Falk Pharma, FAES Pharma, Ferring, Janssen, MSD, Otsuka, Pfizer, Shire, Takeda, Galapagos, Boehringer Ingelheim and Tillots. Alessandro Armuzzi: consulting and/or advisory board fees from AbbVie, Amgen, Biogen, Boehringer Ingelheim, Bristol-Myers Squibb, Celgene, Celltrion, Eli-Lilly, Ferring, Galapagos, Gilead, Janssen, MSD, Mylan, Pfizer, Protagonist Therapeutics, Roche, Samsung Bioepis, Sandoz, Takeda; lecture and/or speaker bureau fees from AbbVie, Amgen, Arena, Biogen, Bristol-Myers Squibb, Eli-Lilly, Ferring, Galapagos, Gilead, Janssen, MSD, Mitsubishi-Tanabe, Novartis, Pfizer, Roche, Sandoz, Samsung Bioepis, Takeda; and research grants from MSD, Pfizer, Takeda and Biogen. The remaining authors declare no competing interests. The remaining Authors declare no competing interests.

Introduction

Crohn's disease is characterized by patchy, transmural inflammation predominantly affecting the ileum and colon. Ileo-colonoscopy remains the gold standard for assessing mucosal inflammatory activity in clinical trials, and can be quantified using the Simple Endoscopic Score for Crohn’s Disease (SES-CD). However, ileo-colonoscopy is invasive, therefore there has been growing interest in non-invasive imaging modalities such as magnetic resonance imaging (MRI). The Simplified Magnetic Resonance Index of Activity (sMaRIA) is an MRI-based score which provides a standardised approach for evaluating disease activity. Most previous studies investigating the correlation between sMaRIA and SES-CD have been single centre studies with limited numbers of readers and performed only in patients with longstanding disease. While MRI is superior to colonoscopy for assessment of transmural penetrating and stricturing disease, the performance of sMaRIA for assessment of mucosal activity in the post-treatment setting remains unanswered.

Aims & Methods

To evaluate the performance of sMaRIA for detection of mucosal inflammation in patients with recently diagnosed Crohn’s disease using ileo-colonoscopy and SES-CD as the reference standard. We conducted a post-hoc analysis of the PROFILE trial, in which patients had undergone paired ileo-colonoscopy and small bowel MRI at end of trial (week 48). Central reading was performed by validated pools of consultant radiologists (MRI) and gastroenterologists (ileo-colonoscopy), blinded to clinical data. Correlations between SES-CD and sMaRIA scores were assessed, alongside classification performance and remission rates by treatment arm. For this study endoscopic and MRI remission were respectively defined as SES-CD ulcer subscore <1 and sMaRIA ulcer subscore of 0.

Results

253 patients had paired ileo-colonoscopy and MRI at end of trial with a median time gap of 22 days between investigations and no treatment change over this period. The overall correlation between SES-CD and sMaRIA scores at week 48 was poor (Spearman’s rank=0.196). [KS(UHNFT1] Overall sMaRIA had low sensitivity (27%) [NN2] for identifying active inflammation evident on ileo-colonoscopy. On segmental analysis MRI detected only 4/146 ulcerated segments (SES-CD ulcer subscore >0). Focusing on patients with ileal involvement, terminal ileal ulcer subscores also showed low correlation (Spearman’s rank=0.11). When using the mucosal component of MRI / sMaRIA to assess ulcer-free remission, no difference in end of trial outcomes was observed for “top-down” vs “accelerated step-up” treatment groups (95% vs 96% remission, p=0.728). Rates of ulcer-free remission assessed by SES-CD were different between treatment groups (86% vs 71% remission, p=0.004).

Conclusion

In this cohort of recently-diagnosed patients with Crohn’s disease and predominantly inflammatory disease phenotype, sMaRIA showed low sensitivity for detecting mucosal inflammation / ulceration. Had sMaRIA been used instead of SES-CD in the PROFILE trial it would have failed to identify the benefit of the “top-down” treatment strategy. These findings suggest that sMaRIA on MRI should not replace SES-CD scoring on ileo-colonoscopy for assessing mucosal inflammatory activity in Crohn’s disease.

References

A biomarker-stratified comparison of top-down versus accelerated step-up treatment strategies for patients with newly diagnosed Crohn's disease (PROFILE): a multicentre, open-label randomised controlled trialNoor, Nurulamin MNoor, Nurulamin et al.
The Lancet Gastroenterology & Hepatology, Volume 9, Issue 5, 415 - 427

Introduction

Capsule endoscopy (CE) is a minimally invasive and essential tool for assessing small bowel disorders, especially Crohn’s disease. Accurate detection of ulcers and erosions is vital for evaluating disease activity, guiding treatment, and monitoring therapeutic response. Although CE has revolutionized lesion detection, its manual interpretation remains time-consuming and subject to interobserver variability. Recent advances in artificial intelligence (AI), particularly convolutional neural networks (CNNs), offer promising solutions by enhancing diagnostic accuracy, reducing variability, and improving workflow efficiency. This study aimed to develop and validate a robust AI model capable of detecting and distinguishing small bowel ulcers and erosions across multiple CE platforms and clinical centers.

Aims & Methods

In a prospective, multicenter study conducted between January 2021 and April 2024, data were collected from centers in Portugal, Spain, and the United States. Two CE devices—PillCamSB3 and Olympus EC-10—were used to analyze 137 anonymized exams. The AI-assisted readings generated by a deep learning model were compared to standard-of-care (SoC) interpretations, with an expert consensus panel serving as the reference standard. Performance metrics included sensitivity, specificity, positive and negative predictive values (PPV, NPV), and area under the receiver operating characteristic curve (AUROC).

Results

Expert consensus identified ulcers and erosions in 56 patients (40.9%). Compared to the reference standard, SoC readings achieved 60.7% sensitivity, 98.8% specificity, 97.1% PPV, 78.4% NPV, and 83.2% accuracy. In contrast, the AI-assisted model achieved 94.6% sensitivity, 80.2% specificity, 76.8% PPV, 95.6% NPV, and 86.1% accuracy. The AI model was found to be both non-inferior (p < 0.001) and statistically superior (p < 0.001) to SoC. Notably, it identified 68 lesions, nearly double those detected by SoC (35), and demonstrated consistent performance across devices and centers. The mean AI-assisted reading time was 239 seconds (SD 138 seconds) per exam.

Conclusion

The AI-assisted model significantly outperformed conventional SoC in the detection of small bowel ulcers and erosions, achieving higher sensitivity and overall accuracy while substantially reducing reading time. This is the first multicenter validation of an interoperable AI solution for CE, capable of high diagnostic performance across different devices and clinical environments. These findings highlight the potential of AI to transform endoscopic evaluation and management in inflammatory bowel disease.