E. Pertsinidou[1,2], B. Salomon[3], D. Bergemalm[4], S. Salihovic[3], C.R.H. Hedin[5,6], M. Ling Lundström[7], Å.V. Keita[8], M.K. Magnusson[9], C. Eriksson[4], M.B. Bengtson[10], T.B. Aabrekk[10,11], R. Movérare[2,12], N. Rydell[2], H. Ekoff[2,7], J. Rönnelid[1], M. D’Amato[13,14,15], T.E. Detlie[11,16], G. Huppertz-Hauss[17], R. Opheim[17,18], P. Ricanek[16,19], V.A. Kristensen[18,20], L. Öhman[9], J.D. Söderholm[8], R. Kruse[3], C.M. Lindqvist[3], M. Carlson[7], D. Repsilber[3], M.L. Høivik[11,18], J. Halfvarson[4]

Affiliates

[1]Uppsala University, Department of Immunology, Genetics and Pathology, Uppsala, Sweden. [2]Thermo Fisher Scientific, Uppsala, Sweden. [3]Örebro University, School of Medical Sciences, Örebro, Sweden. [4]Örebro University, Department of Gastroenterology, Faculty of Medicine and Health, Örebro, Sweden. [5]Karolinska Institutet, Department of Medicine Solna, Stockholm, Sweden. [6]Karolinska University Hospital, Centre for Digestive Health, Department of Gastroenterology, Dermatovenereology and Rheumatology, Stockholm, Sweden. [7]Uppsala University, Department of Medical Sciences: Gastroenterology and Hepatology, Uppsala, Sweden. [8]Linköping University, Department of Biomedical and Clinical Sciences, Linköping, Sweden. [9]University of Gothenburg, Sahlgrenska Academy, Department of Microbiology and Immunology, Institute of Biomedicine, Gothenburg, Sweden. [10]Vestfold Hospital Trust, Department of Gastroenterology, Tønsberg, Norway. [11]University of Oslo, Institute of Clinical Medicine, Oslo, Norway. [12]Uppsala University, Department of Medical Sciences: Respiratory, Allergy and Sleep Research, Uppsala, Sweden. [13]Gastrointestinal Genetics Lab, CIC bioGUNE - BRTA, Derio, Spain. [14]Ikerbasque, Basque Foundation for Science, Bilbao, Spain. [15]Department of Medicine and Surgery, LUM University, Casamassima, Italy. [16]Akershus University Hospital, Department of Gastroenterology, Lørenskog, Norway. [17]University of Oslo, Institute of Health and Society, Oslo, Norway. [18]Oslo University Hospital, Department of Gastroenterology, Oslo, Norway. [19]Lovisenberg Diaconal Hospital, Department of Gastroenterology, Oslo, Norway. [20]Lovisenberg Diaconal Hospital, Unger-Vetlesen Institute, Oslo, Norway.

Abstract

Objective: Autoantibodies against integrin αvβ6 have recently been associated with ulcerative colitis (UC), its preclinical phase, and UC outcomes. However, data on the diagnostic utility of the autoantibody and its dynamics are scarce. Therefore, we aimed to examine the diagnostic capacity of the anti-αvβ6 autoantibody. Moreover, we assessed its utility in defining a specific phenotype of UC and ascertained its temporal dynamics in relation to the disease course.

Methods: Using an in-house assay based on the EliA technology (Thermo Fisher Scientific, Uppsala, Sweden), anti-αvβ6 IgG autoantibodies were measured in serum from adult patients with suspected IBD in the Swedish inception cohort SIC IBD (discovery cohort, n=473) and the Norwegian population-based inception cohort IBSEN III (validation cohort, n=570). A cut-off of 400 UA/l was applied, corresponding to the 96th percentile of 196 healthy individuals. The course of UC was categorized 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, or >2 courses of corticosteroids/cumulative dose of >2.5 g. Inflammatory proteins (n=92) were measured (Olink Proteomics, Uppsala). Welch’s t-test was used with Benjamini-Hochberg correction, applying a 5% false discovery rate (FDR). Linear-mixed models were applied to investigate differences in autoantibody expression between diagnosis and three months follow-up.

Results: The discovery cohort comprised 327 IBD patients and 146 symptomatic controls without any discernible evidence of IBD, and the validation cohort included 366 IBD patients and 204 symptomatic controls. Patients with UC and those with colonic Crohn’s disease in the discovery cohort had higher median anti-αvβ6 levels compared to symptomatic controls. The anti-αvβ6 autoantibody demonstrated a sensitivity of 73% and a specificity of 84% for distinguishing UC from CD and symptomatic controls in the discovery cohort and 79% and 91%, respectively, in the validation cohort.
Anti-αvβ6 positivity was associated with a severe UC phenotype, characterized by more extensive colitis (p=0.008), severe Mayo endoscopic activity (p=0.005), and pronounced systemic inflammation, defined as higher hsCRP (p=0.04) and lower albumin concentrations (p=0.01), in the discovery cohort. However, no association of f-Calprotectin was observed. Higher endoscopic activity in anti-αvβ6 positive UC patients was confirmed in the validation cohort. In both cohorts, an increasing extent of inflammation and endoscopic activity was observed with higher median levels of anti-integrin αvβ6.
Five proteins (SYND1, IL-17A, GZMB, MMP1, CXCL13) were significantly upregulated in anti-αvβ6 positive UC patients in the SIC-IBD, and two of them (IL-17A and GZMB) were validated in the IBSEN III cohort. Compared to patients with indolent UC, patients with an aggressive course presented with higher anti-αvβ6 levels at diagnosis in the discovery cohort (p=0.02) and potentially also in the validation cohort (p=0.13). A significant interaction of disease course was observed in autoantibody levels between diagnosis and 3 months in the SIC cohort (p<0.0001), as the levels decreased in patients with indolent UC but not in those with an aggressive course.

Conclusion: Anti-αvβ6 autoantibodies demonstrate a diagnostic capacity for UC and define a severe phenotype of the disease. Its temporal dynamics differ between patients with an aggressive and an indolent disease course.