DOP06 Histologic activity in Ulcerative Colitis patients achieving endoscopic healing is associated with higher rate of relapse and a distinct mucosal microbial and transcriptional profile
Hernandez-Rocha, C.(1);Turpin, W.(2);Nayeri, S.(3);Borowski, K.(3);Stempak, J.(3);Conner, J.(4);Silverberg, M.S.(1)
(1)Zane Cohen Centre for Digestive Diseases- Lunenfeld-Tanenbaum Research Institute- Sinai Health System, Division of Gastroenterology- Mount Sinai Hospital- University of Toronto, Toronto- ON, Canada;(2)Zane Cohen Centre for Digestive Diseases- Lunenfeld-Tanenbaum Research Institute- Sinai Health System, Department of Medicine- University of Toronto, Toronto- ON, Canada;(3)Zane Cohen Centre for Digestive Diseases- Lunenfeld-Tanenbaum Research Institute, Sinai Health System, Toronto- ON, Canada;(4)Department of Pathology & Laboratory Medicine, Mount Sinai Hospital, Toronto- ON, Canada
Background
Histologic remission (HR) in ulcerative colitis (UC) has been associated with better clinical outcomes. However, this therapeutic target is hard to achieve with the available medications. We aim to assess the relapse-free survival (RFS) of patients with persistent histologic activity (HA) upon achieving endoscopic healing (EH) and the microbial and molecular signatures associated with this condition.
Methods
Left-sided and extensive UC patients and healthy controls (HC) were recruited at colonoscopy and sigmoid colon (SC) biopsy samples were obtained for HA evaluation, 16S rRNA sequencing and host RNA-seq. EH was defined as Mayo endoscopic subscore (MES) 0-1, and HA and histologic remission (HR) as the presence and absence of neutrophil infiltration, respectively. UC relapse was retrospectively evaluated and defined as symptomatic and endoscopic worsening. Cumulative RFS was assessed by log-rank test. Principal component analysis (PCA) was used for capturing cluster structure. For differential expressed genes (DEG) and enrichment pathway analyses, glmQLFTest in EdgeR and enrichKEGG in clusterProfiler were applied, respectively. Genes with log fold change > 2 and a log count per million > 1 were considered DEG. Dada2 algorithm in QIIME2 for amplicon sequence variant generation and ANCOM-BC package for differential abundance (DA) taxa analysis were applied. P-values of omics data were corrected by false discovery rate (FDR) method.
Results
Sixty-five UC patients had EH (37 MES 0 and 28 MES 1). Nine (20.9%) MES 0 and 17 (40.5%) MES 1 showed HA. Over a median follow-up of 4.2 years (IQR 2.6-6.1), UC patients with HA showed a significant lower RFS compared to HR (p < 0.05; Figure 1). This difference was not observed between MES 0 and MES 1 (p = 0.2). Transcriptomics data was available for 27 samples with HR and 9 samples with HA, as well as 43 HC and 13 MES 2-3 samples. PCA showed that HR and HA samples cluster with HC and MES 2-3 samples, respectively (Figure 2). Compared to HR, 163 DEG (118 up and 45 downregulated) were found in HA. Enrichment analysis showed significant upregulation of cytokine-cytokine interaction, Th17 and chemokine signaling pathways (Figure 3). Microbiome data was available for 43 and 19 samples with HR and HA, respectively. Three genera had significant DA in HA compared to HR (Figure 4).
Conclusion
HA in UC is better than MES at distinguishing a higher risk of relapse and has a transcriptomics profile closer to active endoscopic disease despite EH with significant upregulation of chemokine, cytokine and Th17 pathways. A different microbial signature is also observed in HA. These particular signatures could help to define strategies to reduce the risk of UC relapse in patients achieving EH.
