Back

A distinct serum metabolic signature outperforms C-reactive protein as a non-invasive marker for monitoring disease activity in Inflammatory Bowel Disease

Welz, L.; Laabs, B.-H.; Harris, D. M. M.; Schuchardt, S.; Tran, F.; Waschina, S.; Frey, N.; Franke, A.; Verstockt, B.; Vermeire, S.; Rosenstiel, P.; Schreiber, S.; Aden, K.

2025-12-11 gastroenterology
10.64898/2025.12.10.25341951 medRxiv
Show abstract

BackgroundAchieving deep remission, such as e.g. histo-endoscopic remission, is a key objective in inflammatory bowel disease (IBD). However, this metric depends on invasive procedures that are impractical in routine care, whereas current non-invasive blood biomarkers demonstrate only limited correlation with disease activity. ObjectiveWe used serum metabolomics to identify blood-based signatures that comprehensively assess disease activity. DesignSerum was collected from HC (n=195), UC (n=183; Kiel, discovery), UC (n=52; Leuven, validation) and CD (n=98; Kiel) cohorts alongside composite assessment of disease activity using endoscopic, clinical, biochemical, and histopathological criteria. Targeted metabolomics was performed with the Biocrates MxP Quant 500(XL) kit. Statistical analysis involved principal component analysis (PCA), linear mixed models (LMM), generalised estimating equation (GEE) and ML (logistic, LASSO regression). ResultsWe identified numerous serum metabolites predominantly related to amino acid, bile acid and lipid metabolism that differed significantly between HC and inactive or active UC/CD. Among those, increased beta-alanine and Hex2Cer(d18:1/16:0) accompanied by diminished tryptophan and histidine distinguished active from inactive UC as well as HC from active IBD across several disease activity definitions. Combining those metabolites better estimated disease activity than C-reactive protein (CRP). Lastly, prior to initiation of advanced therapy, baseline metabolites and LASSO-selected combinations thereof exhibited limited ability to predict remission. ConclusionSerum metabolomics distinguished IBD from HC and inactive from active disease, with a four-metabolite panel outperforming CRP in comprehensively assessing disease activity. Despite restricted predictive performance, our findings underscore the value of serum metabolomics in elucidating IBD pathophysiology and improving disease monitoring. What is already known on this topicSerum metabolomics reveal systemic immunometabolic alterations in IBD, but prior studies lack composite disease activity measures and independent validation. What this study addsCross-cohort analysis of comprehensively assessed disease activity revealed metabolic signatures that distinguish IBD from HC and inactive from active disease states, with superior discriminatory power compared to CRP. However, baseline metabolites exhibit limited ability to predict outcomes. How this study might affect research, practice or policyOur findings highlight both the potential and current constraints of serum metabolomics for clinical application, emphasizing the need for standardized, longitudinal, and multi-cohort studies to enable reproducible biomarker discovery in IBD.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.