Cure8 research brief
Why This Matters
The work identifies factors that predict loss of response to infliximab and suggests maintenance drug levels are the strongest predictor, which could influence monitoring strategies for people with Crohn's disease on infliximab.
Who Should Pay Attention
Patients on infliximab, gastroenterologists/IBD clinicians, and researchers working on biomarkers or AI prognostic models.
Study Snapshot
What To Know
Researchers retrospectively analyzed 637 Crohn's patients starting infliximab and compared several time-to-event ML models. A gradient boosting machine performed best, with better discrimination at longer follow-up (24–36 months) than at 12 months.
Key predictors included maintenance IFX trough levels, disease duration, induction-phase anti-drug antibodies, platelet count, sex, concomitant immunosuppressant use, and white blood cell count.
The authors used SurvSHAP to make the model more interpretable and found maintenance troughs had the largest effect range; concomitant immunosuppressants showed a protective signal. The study is internally validated but limited to a single-center retrospective cohort, and the authors call for external multicenter validation before clinical implementation.
Keep In Mind
This report is an internally validated, retrospective single-center study (abstract/full-text level). External multicenter validation is needed before applying the model in routine care. The 12-month predictive performance was modest compared with later time points.
Source Details
Review the original publication for the complete reporting, methods, and context.
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.