Magnetic resonance cell size imaging reveals intestinal fibrosis to predict intestinal disease progression in patients with Crohn’s disease

Abstract Background Intestinal fibrosis in Crohn’s disease (CD) is driven by mesenchymal cell activation, resulting in adverse outcomes. We aimed to evaluate the efficacy of time-dependent diffusion magnetic resonance imaging (TD-dMRI) in characterizing fibrosis-associated cellular properties and predicting disease progression in CD. Methods This prospective study enrolled 145 CD patients undergoing TD-dMRI to map fibrotic cellular characteristics (eg, cell diameter [d]). The performance of TD-dMRI was evaluated in surgical cohort 1 (31 patients, 63 specimens) based on myofibroblast/fibroblast area ratio from immunohistochemical staining, and further validated in surgical cohort 2 (21 patients, 25 specimens) using vimentin+ cell diameter from immunofluorescent staining. A follow-up cohort of 93 patients with different baseline mesenchymal cell phenotypes characterized by TD-dMRI was monitored for disease progression. Results TD-dMRI-derived d correlated strongly with myofibroblast/fibroblast area ratio in surgical cohort 1 (r = 0.58; P < .001) and with vimentin+ cell diameter (r = 0.70; P < .001) in surgical cohort 2. Cell diameter d was the most discriminative parameter for distinguishing diseased and normal samples (AUC = 0.86; P < .001), with d ≥ 11 μm indicating profibrotic mesenchymal cell activation state. In all cohorts, d correlated positively with wall thickness and negatively with the narrowest lumen diameter and stenosis index (|r|=0.43–0.51, all P < .001). CD patients with d ≥ 11 μm exhibited higher disease progression rate (33% vs. 7%; P = .008) and shorter disease-progression-free survival (P = .003) than those with d < 11 μm. Moreover, d was the most prominent predictor for disease progression (HR: 1.3; P < .001). Conclusions TD-dMRI-derived d serves as a noninvasive microstructural biomarker for intestinal fibrosis in CD, which significantly enhances the accuracy in predicting disease progression risk.