P016 Reduction of mucosal (active) eosinophils, B cells and T cells after vedolizumab therapy in patients with ulcerative colitis

Background Patients with ulcerative colitis (UC) are often treated with biological therapies or small molecules. Knowledge about the impact of these therapies on the intestinal and peripheral blood immune cell composition is scarce. Therefore, we investigated how advanced therapies modulate immune cell distribution in UC patients. Methods We included 30 UC patients (53% male, median age 42 years) who started a biological or small molecule. Before the first drug administration, mucosal colonic biopsies and a peripheral blood sample were obtained. At the end of induction, colonic biopsies and peripheral blood were sampled again. Patients starting adalimumab (n=2), infliximab (n=3), vedolizumab (n=11), ustekinumab (n=6), ozanimod (n=2) and the JAK inhibitors filgotinib (n=3) and tofacitinib (n=3) were included. Endoscopic improvement was defined as a Mayo endoscopic subscore of 0-1 at the end of induction. From the biopsies, a single-cell suspension was made. Intestinal and circulating immune cells were characterized via flow cytometry. Statistical analysis was performed using a paired t-test. Results Independent of the mechanism of action (MOA), patients responding to therapy showed a decrease of colonic granulocytes (neutrophils (p<0.0001) (Figure 1A), basophils (p<0.0001) (Figure 1B) and eosinophils (p=0.008) (Figure 1C)), active eosinophils (p=0.002) (Figure 1D)), B cells (p=0.05) (Figure 1E), regulatory T cells (p<0.0001) (Figure 1F) and T helper (Th) 2 cells (p=0.02) (Figure 1G), balanced with an increase of Th1 cells (p=0.03) (Figure 1H). In peripheral blood, eosinophils increased in patients not responding to therapy (p=0.05) (Figure 1I). Furthermore, we observed that only patients starting vedolizumab (n=11) showed a decrease in colonic eosinophils (p=0.02) (Figure 1J), active eosinophils (p=0.002) (Figure 1K), B cells (p=0.03) (Figure 1L) and T cells (p=0.004) (Figure 1M). Considering only non-vedolizumab patients (n=19), we did not observe this effect. Conclusion UC patients responding to advanced therapies showed a different intestinal immune cell distribution compared to non-responders, regardless of MOA. Vedolizumab therapy furthermore decreased several mucosal immune cell subsets that migrate to the gut through α4β7-MAdCAM-1 binding. While the effect of vedolizumab on B cells and T cells was previously described, we have now potentially identified an additional eosinophil-reducing effect in the colon.