Benita Salomon[1], Marie Carlson[2], Daniel Bergemalm[3], Charlotte R H Hedin[4,5], Johan D Söderholm[5,6], Åsa V Keita[5], Adam Carstens[8,9], Olof Grip[10], Jan Marsal[10], Carl Eriksson[3,11], Hans Strid[12], Carl Mårten Lindqvist[1], Lena Öhman[13], Maria K Magnusson[13], Mauro D’Amato[14,15,16], the BIO IBD study group, Dirk Repsilber[1], Robert Kruse[1,17,18], Jonas Halfvarson[3]

Affiliates

[1] School of Medical Sciences, Örebro University, Örebro, Sweden[2] Department of Medical Sciences, Gastroenterology Research Group, Uppsala University, Uppsala, Sweden[3] Department of Gastroenterology, Faculty of Medicine and Health, Örebro University, Örebro, Sweden[4] Karolinska Institute, Department of Medicine Solna, Stockholm, Sweden [5] Karolinska University Hospital, Gastroenterology Unit, Department of Gastroenterology, Dermatovenereology and Rheumatology, Stockholm, Sweden [6] Department of Biomedical and Clinical Sciences, Linköping University, Linköping, Sweden.[7] Department of Surgery, Linköping University, Linköping, Sweden[8] Department of Internal Medicine, Ersta Hospital, Stockholm, Sweden.[9] Karolinska Institute, Department of Medicine Huddinge, Stockholm, Sweden. [10] Department of Gastroenterology, Skåne University Hospital, Malmö/Lund, Sweden. [11] Clinical Epidemiology Division, Department of Medicine Solna, Karolinska Institute, Stockholm, Sweden.[12] Department of Internal Medicine, Södra Älvsborgs Hospital, Borås, Sweden.[13] Department of Microbiology and Immunology, Institute of Biomedicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden.[14] Department of Medicine and Surgery, LUM University, Casamassima, Italy.[15] Gastrointestinal Genetics Lab, CIC BioGUNE–BRTA, Derio, Spain.[16] Ikerbasque, Basque Foundation for Science, Bilbao, Spain.[17] Inflammatory Response and Infection Susceptibility Centre, (iRiSC), Faculty of Medicine and Health, Örebro University, Örebro, Sweden.[18] Department of Clinical Research Laboratory, Faculty of Medicine and Health, Örebro University, Örebro, Sweden.

Abstract

Introduction
Better prognostic measures for ulcerative colitis (UC) could significantly advance patient care. While the prognostic capacity of circulating proteins in UC has been explored, the role of mucosal proteins remains largely unknown. We examined mucosal protein markers in patients with incident ulcerative colitis and evaluated their prognostic value.

Methods
Biopsies from macroscopically inflamed colonic/rectal mucosa of adult patients in the Swedish inception cohort of IBD (SIC IBD) were obtained at diagnosis of UC. Patients were followed prospectively, and clinical data were recorded after 3 and 12 months. Disease course was categorised as indolent or aggressive at 12 months, based on a composite outcome of colectomy, hospital admission for active disease, treatment refractoriness towards ≥2 biological agents, and the use of >2 courses of corticosteroids, or a cumulative dose of >2.5 g. Relative estimates of 162 protein markers were assessed in homogenised tissue supernatants, using proximity extension assay technology (Olink Proteomics, Uppsala, Inflammation and Oncology II panel). Mann-Whitney U test, with Benjamini-Hochberg correction was used to identify differentially regulated mucosal proteins in aggressive vs indolent disease course, with a 5% false discovery rate (FDR). Smoothly clipped absolute deviation regularised logistic regression models were used to identify prognostic signatures of distinguishing aggressive from indolent disease course. Performance was estimated in a leave-one-out cross-validation and reported as the area under the receiver operating characteristic (ROC) curve (AUC).

Results
117 patients provided a macroscopically inflamed colonic/rectal biopsy at diagnosis of UC. Basic demographics and clinical characteristics are presented in Table 1. Relative protein levels of WFDC2 and CCL20 were significantly lower in lysates from patients developing an aggressive course vs patients developing an indolent course, while estimates of MMP-1, CCL11, WISP1, OPG, RSPO3 and VEGFR2 were higher (Figure 1A). Regularized logistic regression identified signatures restricted to 28 proteins, distinguishing aggressive from indolent UC courses, yielding an AUC of 0.68 (95% confidence interval (CI): 0.56-0.79) for left-out samples (Figure 1B). Incorporating extent of inflammation at diagnosis in the model improved the AUC to 0.71 (95% CI: 0.60-0.82).

Conclusion
We identified prognostic mucosal protein signatures associated with future course of ulcerative colitis by analysing inflamed mucosal biopsies that were obtained at diagnosis. These protein markers may highlight pathways of relevance for ulcerative colitis outcomes.