Intestinal fibrosis remains a major therapeutic challenge in Crohn’s disease (CD) management. While the brain-gut axis is central to CD pathogenesis, whether neural alterations may provide complementary information about intestinal fibrogenesis remains unexplored. We aimed to develop a multiparametric brain MRI “fingerprint” model to stratify fibrosis severity and to explore potential metabolite-related links between brain and fibrosis. In this two-center prospective study (n = 109), we integrated multiparametric brain MRI (functional, structural, and microstructural), magnetic resonance enterography (MRE) with magnetization transfer (MT), and serum targeted metabolomics (429 metabolites). A brain fingerprint model was developed using machine learning (recursive feature elimination) to differentiate moderate-to-severe fibrosis (BF2) from non-to-mild fibrosis (BF1). Exploratory mediation analyses were performed to assess potential indirect associations among brain imaging features, serum metabolites, and fibrosis-associated MRE findings. The brain fingerprint model, comprising 13 neuroimaging features, demonstrated robust performance in identifying moderate-to-severe fibrosis in CD patients (AUC training =0.923; AUC validation =0.837; AUC test =0.846). The brain fingerprint index positively correlated with both normalized MTR ( r = 0.53) and bowel stricture ( r = 0.31). Patients with moderate-to-severe fibrosis exhibited reduced cortical R2* and Amplitude of Low-Frequency Fluctuations (ALFF), alongside increased activity in the isthmus cingulate cortex. Exploratory mediation analyses yielded 13 nominal indirect associations involving serum metabolites; none met the Benjamini–Hochberg false discovery rate-adjusted significance threshold. Multiparametric brain MRI features were associated with normalized MTR-defined intestinal fibrosis severity in CD. The resulting brain fingerprint demonstrated robust discrimination between MTR-defined fibrosis strata and may provide a noninvasive extraintestinal imaging marker complementary to local bowel imaging.
No takes yet. Share an insight, caveat, or question.
Shen et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: