P319. Inflammatory uncomplicated phenotype at the start of biological therapy predicts best surgery-free outcomes
G. Moran1, G. Kaplan1, H. Yang1, C. Seow1, S. Devlin1, L. Dieleman1, R. Fedorak1, S. Ghosh1, R. Panaccione1
1Alberta IBD Consortium, Canada
Background: The effect of biological therapy on surgical rates in Crohn's disease (CD) is unclear. Our objective is to describe the post-biologic surgical incidence and identify prognostic factors.
Methods: A CD cohort (n = 104) prescribed infliximab or adalimumab between 1/7/2000 and 29/6/2011 was identified. The primary outcome was post-biologic CD-related surgery. The primary exposure of interest was disease behavior at onset of first biologic: inflammatory (B1), fibrostenotic (B2) or penetrating disease (B3). Secondary factors collected were: age at diagnosis; sex; time from diagnosis to onset of first biologic; and pre-biologic surgical history. The surgical rate following the start of the first biologic was calculated assuming a Poisson distribution. Kaplan–Meier (KM) survival curves of post-biologic surgery, stratified by disease behavior at onset of biologics, were compared using the log rank test. Cox proportional regression model evaluated the effect of pre-biologic disease behavior (B1 referent versus B2 and B3) on the post-biologic surgical risk after adjusting for confounders. Risk estimates were presented as hazard rate ratios (HR) with 95% confidence intervals (CI).
Results: 26.0% of CD patients underwent post-biologic surgery (0.09 per person-year; 95% CI 0.06–0.13). B2 (mean time to surgery = 1.6±0.17 years; figure 1) and B3 (mean time to surgery = 1.4±0.08 years) disease behavior showed a significant (p = 0.002) increase in surgical rates when compared to B1 (mean time to surgery = 7.3±0.48 years). B2 (HR = 10.7; 95% CI: 3.3–37.5; p = 0.0001) and B3 (HR = 4.2; 95% CI: 1.2–15.4; p = 0.027) disease behavior at onset of the first biologic showed a significantly increased risk for surgery as compared to B1 after adjusting for age at the onset of biologic (HR = 1.0; 95% CI: 0.97–1.03), pre-biologic surgical status (HR = 0.48; 95% CI: 0.18–1.3), time from diagnosis to onset of biologic (HR = 0.97; 95% CI: 0.91–1.03), and pre-biologic perianal disease (HR = 1.9; 95% CI: 0.75–5.1).
Conclusions: Complicated CD behavior at the start of biological therapy is a significant predictor of surgery. Treating CD at the inflammatory stage may improve disease outcomes.
KM curves indicating the cumulative risk of surgery following onset of first biologic.