Prediction of Postoperative Recurrence in Crohn's Disease

Overview

This multicenter prospective observational cohort study aims to develop and validate a multimodal prediction model for postoperative recurrence in patients with Crohn's disease. Patients with Crohn's disease who undergo intestinal resection will be enrolled and followed after surgery. Residual surgical tissue that would otherwise be discarded, together with routinely collected clinical data and biological samples, will be used for research. Multi-omics analyses, including single-cell transcriptomics and spatial transcriptomics, will be performed to identify biological features associated with postoperative recurrence. Participants with Crohn's disease will be followed for clinical, endoscopic, and imaging recurrence after surgery. The study will use data from a screening cohort to develop a prediction model and a validation cohort to evaluate the performance of the model. A control group consisting of patients undergoing intestinal resection for conditions such as colorectal tumors or diverticular disease will also be included for comparative molecular analyses. No treatment will be assigned by the study protocol. All treatment decisions will be made by the treating physicians according to routine clinical practice.

Crohn's disease is characterized by chronic relapsing intestinal inflammation, and a substantial proportion of patients eventually require intestinal resection because of complications such as strictures, penetrating disease, or failure of medical therapy. Postoperative recurrence remains common, but reliable biomarkers and validated prediction models for identifying patients at high risk of recurrence are still lacking. This is a multicenter, prospective, observational cohort study designed to investigate the predictors and molecular mechanisms of postoperative recurrence in Crohn's disease. The study consists of a screening cohort and a validation cohort. Approximately 200 patients with Crohn's disease undergoing intestinal resection will be enrolled, including 100 participants in the screening cohort and 100 participants in the validation cohort. In addition, approximately 50 patients undergoing intestinal resection for colorectal tumors, diverticular disease, or other non-inflammatory bowel disease conditions will be enrolled as controls. In the screening cohort, residual surgical specimens that would otherwise be discarded will be collected. Clinical characteristics, follow-up information, biological samples, and multi-omics data will also be collected. Molecular profiling will include analyses such as bulk transcriptomics, single-cell transcriptomics, spatial transcriptomics, proteomics, microbiome analysis, and metabolomics, as applicable. Differences between patients with and without postoperative recurrence will be evaluated, and machine-learning methods will be used to construct a multimodal prediction model. The validation cohort will be used to evaluate the predictive performance of the multimodal model in a multicenter setting. Postoperative recurrence will be assessed using clinical, endoscopic, and imaging evaluations during follow-up. The study also aims to investigate molecular pathways associated with postoperative recurrence and to identify potential therapeutic targets. This study is observational. No randomized assignment or protocol-mandated treatment will be provided. Clinical management and postoperative treatment decisions will be made independently by the treating physicians according to routine clinical practice.