Treatment Journey Among Patients With Moderate Ulcerative Colitis in the United States: TARGET-IBD

Introduction: Patients with moderate ulcerative colitis (UC) have a high prevalence of corticosteroid (CS) use despite guidelines and known adverse effects, suggesting unmet needs. We assessed treatment patterns and associations of patient characteristics with the initiation of advanced therapy (AT) among patients with moderate UC. Methods: TARGET-IBD is a non-interventional, longitudinal cohort study of patients receiving care for inflammatory bowel disease at 34 US academic or community gastroenterology sites. Adults with UC diagnosed in 2012 or later, receiving conventional therapy as a first treatment, with no AT initiation (biologic/JAK) within 30 days of conventional start, and Mayo Endoscopic Score (MES)=2 (when unavailable, a “pragmatic” MES definition was used based on presence/severity of inflammation and ulcerations/erosions) were included. A Sankey diagram of the first 3 lines of treatment from start of conventional therapy was generated. Fine and Gray sub-distribution hazard regression models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for AT initiation. Results: Of 1,695 patients with UC, 70 met inclusion criteria. Overall, patients were a median of 36 years, male (51%), non-Hispanic white (73%), had median BMI of 25.9, were majority privately insured (79%), and were receiving care at academic sites (66%). A Sankey diagram of treatment journeys depicted high switching between combinations of 5-ASA, immunomodulator (IMM), and CS following initial treatment (Figure 1). In the multivariable model (Table 1), the likelihood of starting AT was significantly lower for Hispanic or non-white patients compared to non-Hispanic white patients (HR 0.26, 95% CI 0.09-0.77); it was also lower for patients age 40-64 at diagnosis compared to those age 18-39 (HR 0.49, 95% CI 0.23-1.04) and higher for patients with extensive disease (HR 1.20, 95% CI 0.55-2.61) and treated at academic sites compared to community sites (HR 1.81, 95% CI 0.93-3.51), though not statistically significant. Conclusion: Among moderate patients with UC, treatment patterns show high switching between various combinations of 5-ASA, IMM and CS. Even patients with extensive disease did not have a significantly greater likelihood of initiating an AT. These treatment patterns and associations help to characterize patients with moderate UC, an often overlooked population, but also emphasize healthcare disparities that require attention.Figure 1.: Patient Treatment Journeys. Table 1. - Unadjusted and Multivariable-Adjusted Models for Advanced Treatment Initiation Univariable-Unadjusted Multivariable-Adjusted Hazard Ratio (95% CI) P-value Hazard Ratio (95% CI) P-value Age at IBD diagnosis (years) 0.06 0.17 18-39 Reference Reference 40-64 0.43 (0.21, 0.86) 0.49 (0.23, 1.04) >=65 0.71 (0.27, 1.86) 0.68 (0.23, 2.00) Sex 0.26 0.15 Male Reference Reference Female 0.72 (0.41, 1.28) 0.64 (0.35, 1.17) Race/Ethnicity 0.02 0.009 Non-Hispanic White Reference Reference Non-White1 0.27 (0.10, 0.78) 0.26 (0.09, 0.77) Other/Not-reported 0.53 (0.23, 1.24) 0.43 (0.19, 0.97) Site type 0.02 0.08 Community Reference Reference Academic 2.10 (1.10, 3.99) 1.81 (0.93, 3.51) Location of ulcerative colitis2 0.17 0.26 Proctitis Reference Reference Left-sided 1.26 (0.59, 2.67) 0.82 (0.37, 1.81) Extensive 1.85 (0.84, 4.08) 1.20 (0.55, 2.61) Not reported 0.54 (0.15, 1.97) 0.32 (0.08, 1.23) 1Non-white includes non-Hispanic black, Hispanic, and Asian.2Based on earliest reported assessment of UC.Note: Index date for patients diagnosed within 3 years before enrollment was the date of conventional therapy (5-ASA, corticosteroid and/or IMM) start. For patients diagnosed more than 3 years before enrollment, IBD diagnosis date was used as a surrogate for conventional therapy start date.