Yingjie Li,1,* Yang Xie,1,* Kejing Cao,1 Jinyang Zhai,1 Shangjian Luo,1 Xiaoman Liu,1 Qin Pan,1,2 Yingwei Chen1,2 

1Department of Gastroenterology, Xinhua Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, 200092, People’s Republic of China; 2Shanghai Institute of Pediatric Research, Shanghai Key Laboratory of Pediatric Gastroenterology and Nutrition, Xin Hua Hospital, Shanghai Jiao Tong University, School of Medicine, Shanghai, 200092, People’s Republic of China

*These authors contributed equally to this work

Correspondence: Qin Pan; Yingwei Chen, Email pan_qin@yeah.net; chenyingwei@xinhuamed.com.cn

Purpose: Ulcerative colitis (UC) is a chronic inflammatory bowel disease with a relapsing-and-remitting cycle. Mucosal healing is a critical predictor of long-term outcomes; however, the discordance between mucosal healing at different depths and clinical activity complicates UC management.
Patients and Methods: Cell death-related genesets (CDRGs) were analyzed in UC patients, with grouping using unsupervised hierarchical clustering. Concordance between grouping methods was assessed using Cohen’s Kappa and Sankey plots. The predictive value was quantified using the area under the curve. Subgroup analysis explored the additive effects and intrinsic associations between the CDRG-based grouping and other grouping methods. Differentially expressed genes (DEGs) were identified using Wilcoxon test. Robustness was validated in 3 independent cohorts. Single-cell analysis was used to assess cell type-specific impacts. Cytokine expression in colon organoids was measured using RT-qPCR.
Results: CDRGs’ activities effectively predicted the clinical, histological, and endoscopic grouping. CDRG-based grouping emerged as an independent predictor, resolving inconsistence between histological and clinical grouping and improving concordance across clinical, histological, and endoscopic grouping. Integrating endoscopic or histological grouping with CDRG-based grouping improved the predictive value for clinical activity. CDRGs were intrinsically associated with DEGs, impacted non-immune cells, and reshaped immune cell infiltration through intercellular interactions. Epithelial cells were the main target of active CDRGs, exhibiting upregulated MIF, MDK, and LGALS9 in UC. In necroptotic colon organoids, MIF and MDK were significantly upregulated, whereas LGALS9 remained unchanged.
Conclusion: CDRG-based grouping served as an independent predictor to improve the concordance between clinical, histological, and endoscopic grouping in UC, and was of potency to precisely identify UC patients achieving long-term treatment target, namely mucosal healing. Active CDRGs intrinsically associated with genetic and immunological alterations in UC, highlighting their potential as novel biomarkers and therapeutic targets.

Keywords: ulcerative colitis, mucosal healing, cell death, prediction, biomarker, single cell