Search in the Abstract Database

Abstracts Search 2019

P076 Conserved inflammatory profile between mice and humans allow unsupervised patient stratification and temporal allocation of IBD-risk genes

P. Czarnewski1, M. Parigi1, O. Diaz1, S. Das1, C. Sorini1, N. Gagliani1, E. J. Villablanca*1

1Karolinska Institute, Medicine, Stockholm, Sweden

Background

Although ulcerative colitis (UC) patients show heterogeneous clinical manifestation, such as diverse response to biological therapies, they are classified as one group. Therefore, an unsupervised molecular re-classification of UC patients has been evoked to design tailored therapies. Moreover, independently on the re-classification, those UC patients who do not respond to biologicals are in urgent need for novel therapeutic targets. Genome-wide association studies (GWAS) have identified potential new target genes, however, their function and optimal therapeutic window remain to be elucidated.

Methods

Due to unsuccessful attempts to classify UC patients based on their colonic transcriptomic profile, we generated and utilised time-series transcriptome data from a mouse model of colitis, which was then cross-compared with human datasets. We also use the time-series transcriptome mouse data to allocate in time the expression of human IBD-risk genes.

Results

Restricting the analysis to the most differentially expressed genes shared between mouse and human, we were able to cluster UC patients into two subgroups, termed UC1 and UC2. We observed that UC1 transcriptional profile is richer in genes associated with neutrophil activity and cytokine signalling than UC2 transcriptional profile. In addition, only 10% of UC1 patients responded to biological therapies (Figure 1). Finally, we temporally allocate IBD risk genes throughout the different phases of intestinal inflammation—tissue damage and tissue repair—providing useful insights on the time of relevance of the IBD risk genes.

Figure 1. Schematic diagram of the pipeline used to stratify patients in UC1 and UC2.

Conclusion

By forward-translating UC disease information from mouse to human we first identified two molecularly distinct UC subgroups characterised by different immunological signatures and responsiveness to biological therapies and second, we associated IBD risk genes to specific phases during intestinal inflammation and recover. Thus, we proposed a new re-classification of UC patients that might be used in clinical practice accompanied with personalised therapies.