Preclinical Protein Signatures of Crohn's Disease and Ulcerative Colitis: A Nested Case-Control Study Within Large...

Area of Science:

• Biomarker discovery
• Proteomics
• Machine learning in medicine

Background:

• Inflammatory bowel disease (IBD) diagnosis requires reliable biomarkers for early risk identification.
• Current diagnostic methods may not capture preclinical stages of IBD effectively.

Purpose of the Study:

• To identify and validate protein signatures predictive of future inflammatory bowel disease (IBD) diagnosis.
• To assess the potential of these signatures for early detection of Crohn's disease (CD) and ulcerative colitis (UC).

Main Methods:

• Utilized large population-based cohorts (n ≥180,000) with longitudinal follow-up.
• Measured 178 proteins using Olink platforms in blood samples from individuals who later developed IBD and matched controls.
• Applied machine-learning algorithms to identify and validate protein signatures in discovery, validation, inception, and twin cohorts.

Main Results:

• A 29-protein signature differentiated preclinical Crohn's disease (CD) with high accuracy (AUC 0.85-1.0 across cohorts).
• Downregulated proteins related to gut barrier integrity and macrophage function correlated with time to CD diagnosis.
• A distinct protein signature showed predictive ability for ulcerative colitis (UC), with AUCs ranging from 0.67 to 0.95.

Conclusions:

• Identified robust protein signatures for predicting future CD and UC diagnoses, validated across multiple independent cohorts.
• The identified CD signature holds significant potential for the early prediction of the disease.
• Further research may refine these signatures for clinical application in IBD risk stratification.