P117 Risk matrix for prediction of disease progression in a referral cohort of patients with Crohn's Disease
P.L. Lakatos1, N. Sipeki2, I. Altorjay2, G.L. Norman3, Z. Shums3, G. Veres4, P. Antal-Szalmas5, M. Papp*2
1Semmelweis University, 1st Department of Medicine, Budapest, Hungary, 2University of Debrecen, Clinical Center, Institute of Internal Medicine, Department of Gastroenterology, Debrecen, Hungary, 3Inova Diagnostics, Inc., San Diego, United States, 4Semmelweis University, 1st Department of Pediatrics, Budapest, Hungary, 5University of Debrecen, Clinical Center, Department of Laboratory Medicine, Debrecen, Hungary
Early identification of patients with at risk for subsequent complications is essential for adapting treatment strategy. The aim of the present study was to develop a prediction model including clinical and serology markers for assessing the probability of developing advanced disease 3, 5 and 7 years after diagnosis in a prospective referral CD cohort.
271 consecutive CD patients (median follow-up: 10.9 yrs) were included. ASCA IgA,IgG and anti-OMP Plus IgA were determined by ELISA. Detailed clinical phenotypes were determined prospectively from diagnosis during the follow-up by reviewing the patients' medical charts. Analysis was limited to patients with inflammatory disease behaviour at diagnosis. Total exposure to steroids, azathioprine(AZA) or anti-TNFs were 88.2%, 73.8% and 41.7%, respectively. At diagnosis, 45% had ileocolonic disease and 79.7% had inflammatory behaviour, while 52% had complicated disease behaviour and 41.1% had at least one resective surgery at last follow-up. Two definitions were used for advanced disease: 1. having intestinal resection or progression in disease behaviour and 2. having intestinal resection, progression in disease behaviour, or need for thiopurines(IBSEN definition).
ASCA IgA and/or IgG, disease location, and need for early AZA were included in the 5-year prediction matrix. The probabilities of advanced disease during this period varied from 6.2% to 55% depending on the combination of predictors. The 3- and 7-year ASCA-based model resulted in probabilities of advanced disease ranging from 0 to 45.5% and from 11.1% to 64.7%. In addition, the model including ASCA, disease location, and early need for steroids but not age at onset, was only predictive for the outcome at 5-years if the IBSEN definition was used. In contrast, the association was lost if the need for azathioprine was excluded from the advanced disease definition. Similar findings were obtained from in a Cox regression analysis, the combination of ASCA, location and early AZA was associated with the probability to develop advanced disease (pLogRank<0.001).
“Association between ASCA, location, need for early azathioprine and the development of advanced disease”
Original model combining ASCA, early steroids and location, however failed to predict disease progression.
Our prediction models identified substantial differences in the probability of developing advanced disease in the short and intermediate course of CD. Markers identified in this referral cohort were different from those previously published in the population-based cohort suggesting that different prediction models should be used in referral setting.