S902 Intestinal Ultrasound Response and Transmural Remission After 48 Weeks of Treatment With Ustekinumab in Crohn’s Disease: STARDUST Trial Substudy Results by Line of Treatment and by Location

Abstract

Introduction: STARDUST is a phase 3b randomized trial comparing treat-to-target (T2T) strategy vs standard of care (SoC) in patients (pts) with Crohn’s disease (CD) treated with ustekinumab (UST). We report Week (W)48 results from the intestinal ultrasound (IUS) substudy assessing transmural remission (TMREM) by biologic treatment (Tx) history and by most affected bowel segment. Methods: Adult pts with moderate–severe active CD failing previous Tx, received iv, weight-based UST ∼6mg/kg at W0 (baseline [BL]); then sc UST 90mg at W8. At W16, pts with ≥ 70-point reduction in CD activity index were randomized 1:1 to either T2T or SoC arms to receive protocol-specified sc UST. Key IUS endpoints assessed at W4, 8, 16 and 48 vs BL (central reading): IUS response (IUSR; ≥ 25 % bowel wall thickness [BWT] reduction from BL); BWT change from BL (mm); TMREM, BWT normalization; color Doppler imaging signal ≤ 1; normal bowel wall echo-stratification (BWS); absence of inflammatory mesenteric fat (i-Fat). TMREM was assessed using 2 definitions of BWT normalization: predefined (≤ 2.0mm ileum; ≤ 3.0mm colon) and new (≤ 3.0mm overall). Descriptive statistics summarized endpoint dataset as observed. Results: Of 77/88 pts assessed, 31 (40.3 %) were biologic-naïve and the most affected segment at BL was ileum in 64.9 % and colon in 35.1 % of pts. A significant absolute reduction from BL in BWT was seen up to W48: overall, by Tx history from W4, and by location (ileum from W4; colon from W8) (Figure 1). Overall IUSR and TMREM (independent of cutoff) increased progressively up to W48 and were numerically higher for biologic-naïve vs biologic-exposed pts. At W48, normalization of IUS endpoints was more notable for biologic-naïve vs biologic-exposed pts for BWT, BWS and i-Fat, but not vascularization. Overall TMREM (predefined BWT normalization) at W48 was 24.1 % with a clinically relevant difference between colon and ileum (50.0 vs 13.2 %). Using the new cutoff, % of pts with BWT normalization and TMREM almost doubled at W48 for ileum and increased overall with clinically relevant differences between colon and ileum (Table 1). Conclusion: In this first IUS interventional study in CD, biologic-naïve pts treated with UST had a numerically higher IUSR and TMREM vs biologic-exposed pts as early as W4, increasing up to W48. Change in cutoff for ileum seems to impact overall TMREM in a clinically relevant way, but independent of cutoff a numerically higher TMREM with UST was noted in colon vs ileum, increasing progressively up to W48.Figure 1.: Tornado diagram showing main drivers (variables and sensitivity ranges) of the incremental cost-effectiveness ratio (ICER). * Values represent threshold values that reduce the ICER to <$100,000/QALY. Abbreviations: VTE (Venous Thromboembolism), DVT (Deep Vein Thrombosis), PE (Pulmonary Embolism), PTS (Post-Thrombotic Syndrome), WTP (Willingness To Pay), EV (Expected Value)Table 1.: Demographic and alternate dosing characteristics for study cohort