Cure8 research brief
Cure8 research brief
Protein biomarkers that predict IBD years before diagnosis could enable earlier detection, inform risk-stratified monitoring, and reveal early disease pathways that might be targeted to prevent or delay onset.
Researchers working on IBD biomarkers and disease mechanisms, clinicians interested in risk prediction and early detection, and adult patients or caregivers curious about advances in preclinical IBD detection.
This is a preprint reporting an observational study in >52,000 UK Biobank participants with proteomic measurements and no prior IBD.
The team trained machine-learning models to rank nearly 3,000 proteins, selected top candidates, and tested them in Cox survival models; models showed moderate ability to discriminate future IBD cases (best C-index ~0.68 for UC).
The proteomic models performed better than clinical-only models, and smoking remained the only significant clinical predictor when combined with proteins. The study identifies dozens of candidate proteins and highlights potential biological pathways in the preclinical phase, but these findings are exploratory and derived from a single cohort.
External validation, replication in independent datasets, and mechanistic work are needed before clinical use.
Preprint on medRxiv using UK Biobank proteomics and machine learning; not yet peer reviewed. Findings are associative and require external validation and assay standardization before clinical application.
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.