Crohn’s disease (CD) is increasingly recognised as a condition that extends beyond the gastrointestinal tract, with mounting evidence of bidirectional communication between the gut and the brain. While symptoms such as pain, fatigue, and cognitive burden are common, the brain’s response to intestinal inflammation has remained difficult to characterise objectively. A new neuroimaging study now sheds light on this relationship, identifying a distinct brain signature associated with active Crohn’s disease using advanced imaging and machine learning techniques.
The study aimed to determine whether integrated structural and functional brain metrics could serve as biomarkers of disease activity, potentially offering a novel, non-invasive complement to existing clinical and biochemical assessments.
Structural Changes Signal Active Disease
Researchers analysed MRI data from 235 participants across two cohorts, including healthy controls, patients with Crohn’s disease in remission, and those with active disease. Structural brain changes were assessed using voxel-based morphometry, while functional alterations were examined through resting-state functional connectivity analyses.
Compared with healthy controls, patients with Crohn’s disease showed widespread alterations in both brain structure and connectivity. Most strikingly, the gray matter volume of the right inferior frontal gyrus, spanning the opercular and triangular regions, was significantly increased in patients with active disease compared with those in remission. This structural enlargement was positively correlated with fecal calprotectin levels, a well-established marker of intestinal inflammation, directly linking brain changes to gut disease activity.
Network Dysconnectivity and Clinical Correlation
Functional MRI analyses revealed reduced connectivity between the right inferior frontal gyrus and bilateral putamen with key brain networks, including the default mode and sensorimotor networks. These disruptions were observed across patients with Crohn’s disease but were more pronounced in those with active disease. Importantly, the degree of functional connectivity loss correlated with clinical measures of disease activity, suggesting a meaningful relationship between brain network integrity and inflammatory burden.
To assess the diagnostic potential of these findings, the researchers applied a machine-learning pipeline combining Lasso regression and support vector machines. Using ten selected imaging features, including gray matter volume of the right inferior frontal gyrus, the resulting classifier successfully distinguished active from remissive Crohn’s disease. The model achieved an area under the curve of 0.85 in internal testing and 0.73 in an independent validation cohort, indicating robust and generalisable performance.
Toward Objective Biomarkers in Crohn’s Disease
Together, these findings identify a neuroimaging signature of active Crohn’s disease characterised by focal structural enlargement and network-level dysconnectivity. By correlating closely with established markers of gut inflammation, this brain-based signature shows promise as an objective biomarker for disease stratification. If validated further, such approaches could enhance understanding of the brain–gut axis and support more precise monitoring of disease activity in Crohn’s disease.
Reference
Bao C et al. Integrated structural and functional brain imaging reveals biomarkers of disease activity in Crohn’s disease. BMC Gastroenterol. 2025;DOI: 10.1186/s12876-025-04547-x.
