Inflammatory bowel diseases
Cure8 research brief
Why This Matters
Predicting who will respond to anti-TNF drugs could help avoid ineffective treatments, reduce delays to effective therapy, and personalize care for children with Crohn's disease.
Who Should Pay Attention
Pediatric IBD clinicians, radiologists, researchers in imaging/AI and IBD, caregivers of children starting biologics, and teams involved in treatment decision-making for patients on anti‑TNF therapy.
Study Snapshot
Story typeResearch paper
Evidence typeResearch paper
Source depthJournal abstract
What To Know
This retrospective study of 92 pediatric patients trained a multi-modal ensemble model that combined radiologist MRE assessment, radiomic features, deep-learning image features, and clinical data. The combined model had stronger discrimination (AUROC ~0.82 when MRE and clinical features were combined) than models using single feature types.
Radiomic texture and intensity features were among the most predictive inputs. The study used MRE performed within 3 months before anti-TNF start and defined response as mucosal healing within 36 months; nonresponders included patients who did not heal, needed surgery, or switched therapy.
This is an early-stage, single-cohort retrospective study and the results require external validation in larger, prospective datasets before clinical use.
Keep In Mind
Retrospective single-cohort study (n=92) with model performance reported as AUROC; results need external validation and prospective testing before clinical implementation.
Source Details
Review the original publication for the complete reporting, methods, and context.
Research paper
Evidence type derived from source or registry metadata.
PublicationInflammatory bowel diseases
AuthorsLi H, Lu Z, Denson N +9 more
Study typeJournal article
Indexed viaEurope PMC
Source typeResearch paper
PublishedAug 7, 2026, 12:00 AM
Content availableJournal abstract
Funding disclosed by the source: the National Institutes of Health - R01 EB030582; the Helmsley Charitable Trust; Cincinnati Children's Research Foundation Academic and Research Committee Award; the National Institutes of Health - P30 DK078392
This Cure8 brief is based on source text from the linked article. Cure8 is informational only and is not a substitute for professional medical advice, diagnosis, or treatment.
The ImproveCareNow Registry: Prediction of remission rates in paediatric Crohn's disease.
Patients, families, and clinicians can use registry-based predictors to better understand which children with Crohn’s disease may be at higher risk of slower or absent remission, which could inform monitoring and...
Why anti-TNF drugs don't work for some kids with Crohn's disease - Medical Xpress
This study suggests a common genetic marker (HLA‑DQA1*05) may help explain why some children stop responding to anti‑TNF drugs and could eventually guide who might benefit from adding methotrexate or closer monitoring.
Pharmacometric Targets During Anti-TNF Induction and Their Association with Deep Remission in Pediatric Crohn’s Disease
The study identifies specific induction exposure targets and biomarkers tied to higher chances of deep remission at 12 months in children on anti-TNF therapy, which could inform earlier dose optimization to improve...