P0488Fibrosis-related serum proteomics to predict response to anti-TNF therapy in pediatric Crohn’s disease: a pilot study

Background Pediatric Crohn’s disease (CD) is often extensive and intensive, predisposing to strictures governed by extracellular-matrix (ECM) remodeling. Consistent with findings from several adult studies, fibrosis-associated biomarkers have been proposed as predictors of therapeutic response to anti-TNF agents, though this relationship has not been tested in children. Early identification of children at risk for fibrostenotic disease could support individualized therapy. We aimed to explore proteomic profiling as a novel approach for detecting circulating biomarkers reflective of fibrogenic pathways, in an attempt to correlate protein expression with the development of fibrosis. Methods In this prospective inception cohort study, children with CD were enrolled at diagnosis or upon initiation of first biologic agent, with follow-up assessment at 4 and 12 months. At all visits, serum samples and clinical data were collected for a fibrosis-focused protein panel that included 18 proteins, including ECM and collagen turnover biomarkers. We limited the current analysis to serum samples from children who initiated anti-TNF as their first biologic therapy. As there is currently no gold standard for identifying non-stenotic fibrosis in CD, and our 12-month follow-up was insufficient for outcomes such as strictures or surgery, we performed a pilot analysis exploring associations with 12-month steroid-free remission (SFR), as defined by the weighted Pediatric CD Activity Index (wPCDAI). Individual protein levels and clinical factors were compared between remission and non-remission groups. Multivariable models were employed to explore predictors of remission. Results Of 39 included children (mean age 13.2 ± 3.9 years, median disease duration 11 months [IQR 3-25]), 21 (54%) achieved SFR at 12 months. First biologic agent included anti-TNF (90%) and other agents (10%). At the 12 month follow up, 12/16 (75%) children for whom stool was available achieved calprotectin of < 250, and new strictures had occurred in 2 (5.1%). None of the examined proteins demonstrated a significant association with 12-month SFR. In multivariable analysis, none of the protein combinations were independently associated with remission. Conclusion Individual fibrosis-associated serum proteins were not predictive of 12-month SFR. The consistent lack of signal across analyses suggests that fibrosis-related serum markers were not major determinants of clinically defined treatment response in this cohort. Future analyses should incorporate endoscopy, imaging and additional longer-term outcomes specifically reflecting fibrosis as well as composite expression patterns to search for predictive potential. Conflict of interest: Ms. Burstyn, Amalia: No conflict of interest Mortensen, Joachim: full time employee and shareholder at Nordic Bioscience Lujan, Rona: No conflict of interest Ledder, Oren: No conflict of interest Orlansky Meyer, Esther: No conflict of interest Yogev, Dotan: No conflict of interest Shteyer, Eyal: No conflict of interest Shamasneh, Ibrahim: No conflict of interest Yuval Bar-Asher, Shira: No conflict of interest Gotesdyner, Leora: No conflict of interest Shawar, Muhammed: No conflict of interest Turner, Dan: No conflict of interest Lev Zion, Raffi: No conflict of interest