DOP78 Efficacy of ozanimod by patient response trajectory subgroups identified using group-based trajectory modelling: a post hoc analysis of the phase 3 True North study

Abstract

Abstract Background Ozanimod (OZA) is approved for the treatment of adults with moderately to severely active ulcerative colitis (UC). A previous post hoc analysis of the phase 3 True North (TN) trial identified 5 discrete groups of patient (pt)-response trajectories to OZA using group-based trajectory modelling (GBTM). The current analysis further evaluated symptomatic, clinical, and mucosal outcomes in these groups. Methods Five trajectory groups were previously identified by GBTM using change from baseline (BL) in partial Mayo score in pts on continuous OZA treatment until W52 (OZA/OZA; N=229): super-response (Group 1; n=38), sustained improvement (Group 2; n=85), partial improvement (Group 3; n=60), fast rebound (Group 4; n=25), and slow rebound (Group 5; n=21) (Schreiber et al. Am J Gastroenterol. 2023;118:S671–2). In each of these groups, symptomatic remission and response were evaluated through W52, and results were compared with placebo (PBO). Clinical and mucosal EPs were evaluated at W10 and/or W52 using nonresponder imputation analysis. Results Significantly more pts in Group 1 vs PBO achieved early symptomatic response on Day (D) 5 (34.2% vs 11.1%; P=0.0002) and remission on D6 (10.5% vs 2.8%; P=0.0264). Symptomatic EPs were sustained for 52 wk in Groups 1–3, whereas Groups 4 and 5 started losing symptomatic efficacy by W18 (Figure 1). The highest proportions of symptomatic remission at W10 were achieved by pts in Groups 1 and 2 (97.4% and 75.3%) and were sustained until W52 (73.7% and 64.7%) (Figure 1A); similarly ~100% of pts in Groups 1 and 2 achieved symptomatic response by W10 and it was sustained in >80% of pts until W52 (Figure 1B). Generally, Groups 1>2>3 had sustained or increased rates of clinical and mucosal EPs from W10 to W52, whereas Groups 4 and 5 had decreased rates of clinical and mucosal EPs at W52 (Table). Corticosteroid (CS)-free remission at W52 was greatest in Groups 1>2>3. Analysis of BL parameters showed that Group 1 had the least prior CS (60.5%), immunomodulator (26.3%), and anti–tumor necrosis factor (TNF) use (23.7%). Conclusion This analysis shows 3 distinctly different OZA populations within the responders, with the super-response Group 1 achieving symptomatic response as early as D5. Early response was related to long-term benefits, with Group 1 (and to a lesser extent with Groups 2 and 3) being more likely to achieve disease control at W52 with OZA. These findings demonstrate the utility of personalized medicine to direct therapeutic choice but need to be confirmed in future prospective cohorts.