Longitudinal Clustering in Crohn's Disease | Nathan Constantine-Cooke

Longitudinal Clustering in Crohn's Disease

Clustering Crohn's disease patients via longitudinal inflammatory marker data

Our current ability to characterise heterogeneity in Crohn’s disease is based on surgical outcomes, hospitalisations, and Montreal disease progression.

In this study, we used longitudinal trajectories of faecal calprotectin, a biomarker for gastrointestinal inflammation, to describe heterogeneity in Crohn’s disease.

We found four clusters of Crohn’s disease patients. Cluster membership was associated with early biologic treatment, smoking status at diagnosis, and upper gastrointestinal involvement. The largest of these clusters demonstrated consistently high faecal calprotectin and were less likely to have received early biologic treatment.

Crohn’s disease patients appear to belong to distinct subgroups based on faecal calprotectin profiles. Instead of only utilising clinical endpoints, considering faecal calprotectin over time allows an inflammation-driven approach to classification and may enable risk stratification in the future.

Manuscript: https://doi.org/10.1016/j.cgh.2023.03.026