P123 Significance of serological markers in the disease course of ulcerative colitis
G. Kovacs*1, N. Sipeki1, K. Palatka1, I. Altorjay1, K. Fechner2, G. L. Norman3, Z. Shums3, G. Veres4, P. L. Lakatos5, M. Papp1
1University of Debrecen, Clinical Centre, Department of Internal Medicine, Division of Gastroenterology, Debrecen, Hungary, 2Institute of Experimental Immunology, Euroimmun AG, Lubeck, Germany, 3Inova Diagnostics, Inc., San Diego, California, United States, 4Semmelweis University, First Department of Paediatrics, Budapest, Hungary, 5Semmelweis University, First Department of Medicine, Budapest, Hungary
Data are few and conflicting regarding the association of serological markers to the disease behaviour, medical treatment, and response to therapy in patients with ulcerative colitis (UC). We aimed to determine the prognostic potential of serological antibodies regarding long-term disease course of an adult prospective UC patient cohort. The association between serological markers and requirements for immunosuppressant or anti-TNF therapy was also evaluated.
Studied were 187 consecutive patients (male 46.0%, median age 40 years, and extensive colitis 33.3%) from a single referral IBD centre. Sera were tested for a panel of different IgA/IgG type autoantibodies (anti-neutrophil cytoplasmic [ANCA], anti-lactoferrin [aLFS], anti-goblet cell, and anti-pancreatic [anti-GP2 and anti-CUZD1] antibodies) by IIFT and for anti-microbial antibodies (ASCA IgG/IgA and anti-OMPPlusTM IgA) using ELISA. Clinical data were available on unfavourable disease outcome, as well as disease activity and medical treatment during the prospective follow-up (median 104 months).
In the study, 73.6%, 62.4%, and 11.2% of UC patients were positive for IgA/IgG type of pANCA, aLFS, and anti-goblet cell antibodies, respectively. Both type of anti-pancreatic antibody occurred in 9% of the patients, whereas ASCA and anti-OMP occurred in 17.7% and 19.8%. Serological antibody status was stable over time. There was no significant association between antibody positivity and gender, age at onset or disease extent. Presence of certain antibodies was negatively associated to the occurrence of extraintestinal manifestations: aLFS IgA/IgG to current smoking status (OR 0.26, p = 0.01) and ocular disease (OR 0.16, p < 0.01), whereas pANCA IgA/IgG to the arthritis (OR 0.36, p = 0.026). During the follow-up period, UC-related hospitalisation occurred in 34.2%, and requirement for colectomy was 3.7%. Exposure to steroids, azathioprine, or anti-TNFs was 77.0%, 37.4%, and 13.4%, respectively. IgA type ASCA and anti-CUZD1 antibody but not other serological markers were associated to an increased likelihood of requirements for immunosuppressant in Kaplan–Meier analysis (pLogRank < 0.01 for both); however, only ASCA IgA was identified as an independent predictor in multivariate Cox-regression model (HR 2.74, 95% CI 1.46–5.14, p < 0.01) comprising age at onset, gender, and disease extent as covariates. At the same time, clinical factors, such as extensive colitis and male gender, but not serological markers, were associated with UC-related hospitalisation, HR 1.8 (95% CI 1.09–2.95, p = 0.019) and HR 6.7 (95% CI 1.6–27.9, p < 0.01), respectively.
The present prospective study displays the limited role of serologic markers in the prediction of disease course in UC.