W1294 Predictors of Primary Non-Response to Infliximab and Optimization in Ulcerative Colitis: A Retrospective Multicenter Study

Abstract

< .0001). Colectomy was performed in 36 patients (18.8%). The probability of maintaining colectomy-free survival was 83.5% (±3.0%), 77.8% (±3.6%), 75.7% (±4.1%), and 61.4% (±10.6%) at 1, 2, 3 and 8 years, respectively. Cox proportional-hazards regression identified three predictors of colectomy: infliximab episodic regimen (HR = 2.41, P = .02), anterior use of cyclosporine (HR = 3.03, P = .002), and absence of short-term clinical response (HR = 6.75, P < .0001). The cumulative probability of first hospitalization for UC flare after infliximab initiation was 33.3% (±3.7%), 39.8% (±4.1%), 42.9% (±4.4%), and 55.4% (±8.27%) at 1, 2, 3 and 8 years, respectively. Cox proportional-hazards regression identified three predictors of first hospitalization for UC flare after infliximab initiation: disease duration ≤ 50 months (HR = 2.14, P = .006), infliximab initiation for acute severe colitis (HR = 3.13, P = .0005), and absence of short-term clinical response (HR = 3.87, P < .0001). CONCLUSIONS: The cumulative risk of infliximab failure exceeded 50% at 3 years among initial responders. About one-fifth of patients underwent colectomy during follow-up, and the cumulative risk of UC flare-related hospitalization was about 40% at 3 years. The risk factors identified in our study may be used in disease modification trials.