Siyan Cao, MD, PhD (he/him/his)
Assistant Professor of Medicine
Washington University School of Medicine
St. Louis, Missouri
Kaiming Ma, BS
Washington University School of Medicine
St. Louis, Missouri
Parakkal Deepak, MBBS, MS
ASSOCIATE PROFESSOR OF MEDICINE, DIRECTOR OF CLINICAL AND TRANSLATIONAL RESEARCH
Washington University School of Medicine
St. Louis, Missouri
Background &
Aims: The interleukin (IL)-23 signaling plays a crucial role in the pathogenesis of Crohn’s disease (CD). IL-23 activates T-helper (Th) 17 cells, IL-17-secreting CD8 T (Tc)17 cells, gd T cells, nature killer (NK) T cells, and group 3 innate lymphoid cells (ILC3) to secrete proinflammatory cytokines including IL-17, IFN-g, and TNF-a. Risankizumab is the first selective IL-23 antagonist approved for moderately to severely active CD. However, there is currently no biomarker to predict response to anti-IL-23 antibodies in IBD. Here, we characterized the cell type- and cytokine-specific pathways in IL-23-activated cells that predict response to risankizumab in CD patients.
Methods: Adult patients with active ileal, colonic, or ileocolonic CD were included (n = 28). Blood samples were collected within 3 months before and 4, 8, 12, and 52 weeks after the initiation of risankizumab. Peripheral blood mononuclear cells (PBMC) were isolated and cryopreserved for batch analysis using mass cytometry (CyTOF). Pre-treatment PBMCs were stimulated ex vivo with IL-23, and stained for lineage markers to identify T-helper (Th)17 cells, IL-17 expressing CD8 T (Tc)17 cells, gd T cells, NKT cells, and mucosal-associated invariant T (MAIT) cells, as well as intracellular cytokines including IL-17, IFN-g, TNF-a, IL-22, IL-6, and GM-CSF, using a pre-optimized and validated antibody panel. Patients’ responses were evaluated 12-40 weeks after the initiation of risankizumab based on clinical symptoms and endoscopic/radiographic assessment.
Results: Among the IL-23-activated cells, increased frequency of NKT cells was identified in the peripheral blood of non-responders before initiation of risankizumab (data not shown). In pre-risankizumab PBMCs, we identified significantly elevated TNF-a in Th17 cells, Tc17 cells, NKT cells, and MAIT cells in non-responders vs. responders (Figure 1). Similarly, in PBMCs 4 weeks after 1st risankizumab infusion, Tc17 cells, gd T cells, NKT cells, and MAIT cells in non-responders produced significantly more TNF-a compared to those cells from responders. In addition, we observed increased IL-17A in multiple IL-23-activated cells in non-responders 4 weeks after starting risankizumab (Figure 2). A similar correlation between therapeutic response and expression of TNF-a or IL-17A was not observed in ustekinumab, an anti-IL-12/IL-23 antibody (data not shown).
Conclusions: CyTOF of patients’ PBMCs before and after starting risankizumab revealed previously unknown molecular signatures, especially augmented TNF-a production in multiple IL-23-activated cells, which correlate with non-response to risankizumab in CD patients. Those findings may help identify response biomarkers for IL-23 antagonists and provide a rationale for using anti-IL-23/anti-TNF combination therapy in a subset of patients with refractory CD.