DOP008Expanded Primary Analysis of the Randomized, Wait-List Controlled STOMP2 Trial Evaluating AVB-114 in Persistent Crohn’s Perianal Fistulas

Abstract Background The STOMP2 trial sought to determine whether the implantable autologous cell therapy AVB-114 is an effective add-on therapy for patients with persistent Crohn’s perianal fistulas. The trial design incorporated wait-list controlled (WLC) elements, allowing eligible control group subjects who completed the 9 mo primary endpoint visit in Part 1 to request AVB-114 treatment in Part 2 and strengthen the evaluation of AVB-114’s safety and effectiveness. Methods Eligible patients had a single perianal fistula tract and had failed prior treatment or had documented medication intolerance to biologic or conventional Crohn’s disease therapy. Subjects at 14 U.S. sites were randomized 1:1 to either standard of care (SoC; fistula optimization including seton placement) or AVB-114 implantation. The primary endpoint was combined fistula remission at 9 mo, defined as no collections 2cm on MRI, visual closure of the external opening, and no drainage of fluid despite gentle finger compression. A logistic regression time from exposure model through 9 mo utilized complete Part 1 data and all available Part 2 data to estimate combined AVB-114 effectiveness. Results Part 1 combined remission (using Non-Responder Imputation) was 8.3% (2/24) for SoC and 45.8% (11/24) for AVB-114 (38% difference, 95% CI 11 – 60%, post-hoc p-value of 0.0078). Twenty SoC subjects were treated with AVB-114 in Part 2. Four of the 9 Part 2 subjects with 9 mo combined remission data available were in combined fistula remission (44.4%). No subjects failed 9 mo radiological remission (first component of the primary endpoint), thus clinical remission (composite of the latter two components) was equivalent to combined remission. AVB-114 clinical remission in Part 1 and Part 2 to-date were similar, and when combined provided narrower estimates with the same coverage (Figure 1). Mean (SD) 9 mo perianal disease activity index (PDAI) scores for SoC, Part 1 AVB-114, and Part 2 AVB-114 were 5.0 (2.6), 4.1 (2.9), and 3.8 (3.5), respectively. Lower 9 mo scores and decreases from baseline were independently associated with an increased probability of clinical remission. Two serious treatment-emergent adverse events (TEAEs; pseudopolyposis and proctalgia) occurred, with the latter attributed to AVB-114. Non-serious product related TEAE were predominantly related to postoperative perianal pain. Two grade 2 (moderate) AEs of cellulitis and abdominal wall abscess were attributed to adipose tissue collection. Conclusion Both the randomized and combined (all-treated) analyses demonstrated substantial AVB-114 effectiveness through 9 mo in this challenging patient population. AEs were anticipated, and generally mild and infrequent. Conflict of interest: Schwartz, David: 1) Abbvie: consultant 2) Janssen: Consultant 3) Takeda: Consultant 4) Avbios- consultant 5) Olympus- consultant Dozois, Eric: TBD Ehman, Eric: TBD Faubion, William: TBD Hudesman, David: Consultant: Avalo, Abbvie, Abivax, Biocon, BMS, CorEvitas, Eli Lilly, Johnson and Johnson, Pfizer, Prometheus, Sanofi, Takeda, Fresenius Kabi Research Support: Pfizer and Johnson and Johnson Espen, Benjamin: employee of Avobis Bio LLC Pabla, Baldeep: Astellas, J & J - Advisory Board Prometheus - Speaker Eli Lilly - Speaker Papadakis, Konstantinos: TBD