P005 Molecular-based classification of ulcerative colitis and its dynamic

Seyed Tabib, N.S.(1)*;Verstockt, S.(1);Verstockt, B.(1,2);Ferrante, M.(1,2);Sabino, J.(1,2);Vermeire, S.(1,2);

(1)KU Leuven, Department of Clinical and Experimental Medicine CHROMETA- Translational Research in GastroIntestinal Disorders TARGID, Leuven, Belgium;(2)University Hospitals Leuven, Department of Gastroenterology and Hepatology, Leuven, Belgium;

Background

A better characterization of inflammatory bowel disease is needed. We previously developed a molecular classification tool for Crohn’s disease, based on a combination of metagenomics (dysbiosis score - MDI), transcriptomics (barrier integrity score - BIS, autophagy score – ATS, unfolded protein response score - UPRS), serum proteomics (inflammation score - IPS) and genetics (polygenic risk score - GRS)1. We here investigated if this score could also apply for ulcerative colitis (UC) and how the individual components correlate with clinical outcomes.

Methods

For a total of 173 UC patients and 181 controls (CO), we collected faecal and blood samples, as well as colon biopsies (Table). Faecal microbiota 16S-sequencing, intestinal RNA-sequencing, and serum proteomics (OLINK inflammatory panel) were performed. Using proteomics and RNA expression data, IPS, ATS, UPRS and BIS were calculated. MDI was calculated as previously described1.  Based on these findings, patients were categorized into quartiles, ranging from Q1 (the least dysfunctional state) to Q4 (the most dysfunctional state). The individual scores were correlated with time to first biological or biological switch, as well as correlations with C-reactive protein (CRP), faecal calprotectin (FC), and the Montreal classification.  The therapeutic value of MDI score was studied in a small dietary intervention pilot study (n=8)2.

Results

None of the individual components of the score were associated with disease extent. Patients diagnosed at a later age had increased ATS and BIS levels. BIS and UPRS showed a positive correlation with disease duration (r=0.36, r=0.54 respectively, p<0.001) (Figure2). MDI (spearman r=0.27, r=0.44, r=0.46 respectively, all p<0.0001) and IPS (r=0.5, r=0.7, r=0.6 respectively, all p<0.0001) correlated with CRP and FC levels, and time to first biologic. UPRS, ATS, MDI, BIS, and IPS were considerably higher in patients who required a switch in biologic therapy (all p<0.001). Finally, we observed a decreasing trend in MDI in response to dietary modification (Figure3).


Conclusion

In UC patients, we created a multifaceted scoring system to molecularly describe disease by the degree of dysbiosis, dysregulation of the immune proteome, and intestinal barrier integrity. We also demonstrated the dynamic of MDI of this score in relation to dietary intervention. Therefore, molecular profiling of patients may represent a new, individualized strategy to the management of UC.

References
1. OP30 ECCO 2020, JCC 14(Supplement_1):S028-S0302. 
2. P767 ECCO 2017, JCC 11(suppl_1):S473-S473.