P502 Predictors of bad response to infliximab in Ulcerative Colitis patients (ECIA study. ACAD)
L. Fernández-Salazar*1, N. Fernández2, R. Sánchez-Ocaña3, D. Joao2, F. Santos3, A. Bouhmidi4, J. Legido5, V. Prieto6, M. Rivero7, R. Pajares8, C. Muñoz9, M. Herranz10, A. Macho1, J. Barrio3, F. Muñoz2
1Hospital Clínico Universitario, Gastroenterology, Valladolid, Spain, 2Complejo Asistencial Universitario, Gastroenterology, León, Spain, 3Hospital Universitario Río Hortega, Gastroenterology, Valladolid, Spain, 4Hospital Santa Bárbara de Puertollano, Gastroenterology, Ciudad Real, Spain, 5Complejo Asistencial de Segovia, Gastroenterology, Segovia, Spain, 6Complejo Asistencial Universitario, Gastroenterology, Salamanca, Spain, 7Hospital Universitario Marqués de Valdecilla, Gastroenterology, Santander, Spain, 8Hospital Infanta Sofía, Gastroenterology, San Sebastián de los Reyes. Madrid, Spain, 9Hospital Virgen de la Salud, Gastroenterology, Toledo, Spain, 10Complejo Asistencial de Ávila, Gastroenterology, Avila, Spain
AntiTNF use is one of the last options of medical ulcerative colitis (UC) treatment before colectomy. ACT studies demonstrated its efficacy. Costs and side effects are disadvantages. We want to identificate: 1. predictor factors for stopping infliximab (IFX) treatment because IFX failure and 2. predictor factors for colectomy.
Retrospective and multicenter study including every UC patient treated with IFX in 10 Spanish centers from June 2003 to May 2014. 187 patients were included. Time from IFX induction to data extraction was 43 ± 27 months. Included variables were related to UC extensión, severity, treatments before IFX, INM intolerance, INM resistance, IFX indication, concomitant use of immunomodulators (IFX+INM), IFX response, intensification, IFX interruption, date and reason, colectomy and colectomy date. Binary logistic regression analyses and Cox regression analyses were used for predictor of both variables: IFX interruption because insufficient response and colectomy. Results were expressed as odds ratios and hazard ratios.
26% of patients had IFX treatment stopped because no or loss of response. 16% of patients underwent colectomy. Independent predictor factors for IFX interuption because an insufficient response were: INM resistance (OR 2,899 IC 1,166-7,212 p 0,022), previous use of leukocitapheresys (LA) anytime before IFX (OR 3,322 IC 1,172-9,416 p 0,024), use of tacrolimus o cyclosporine (TC) anytime before IFX (OR 2,542 IC 0,950-6,802 p 0,063) and INM+IFX (OR 0,342 IC 0,121-0,963 P 0,022). Cox regression model demonstrated that corticosteroid use during IFX induction (HR 1,973 IC 1,002-3,887 p 0,049) and INM+IFX (HR 0,415 IC 0,222-0,777 p 0,006) were independent predictors of IFX interruption because insufficient response. Predictor factors fo colectomy were LA treatment anytime before IFX (OR 3,001 IC 1,075-8,379 p 0,036), severe cortico-resistant flare (OR 2,408 IC 0,952-6,092 p 0,061) and INM+IFX (OR 0,342 IC 0,121-0,963 P 0,022). Cox regression model demontraated severe cortico-resistant flare (HR 2,522 IC 1,080-5,880 p 0,032) and INM+IFX (HR 0,355 IC 0,159-0,792 p 0,011) were independent predictors of colectomy.
LA and TC treatments anytime before IFX, corticoisteroid treatment during IFX induction, severe cortico-resistant flare and IFX monotherapy predict an interuption of IFX treatment because of no or loss of response and colectomy.