ABSTRACT
Introduction Treatment of perianal fistulizing Crohn's disease (PFCD) is a major unmet need. Filgotinib (FIL) is a oncedaily, oral, preferential JAK1 inhibitor in development as a CD treatment. The efficacy and safety of FIL for the treatment of PFCD was evaluated in the phase 2, double-blind, randomized, placebo (PBO)-controlled DIVERGENCE2 study (NCT03077412). Methods Patients 18-75years with PFCD (documented diagnosis of CD for >3months and 1-3 external openings [EOs] with drainage [spontaneous or on compression] for ≥4weeks before screening) previously treated with antibiotics, immunomodulators and/or TNFi were randomized (2:2:1) to receive FIL 200mg, 100mg or PBO once daily for <24weeks. Active luminal CD was permitted providing that CDAI score was ≤300 at screening. The primary endpoint was combined fistula response (reduction of ≥1 from baseline in the number of draining EOs determined by investigator assessment and no fluid collections 1cm on centrally read pelvic MRI) at Week24. Combined fistula remission (closure of all draining EOs present at baseline and no fluid collections >1cm) at Week24 was a key secondary endpoint. The study was not powered for statistical comparisons and was prematurely terminated owing to low recruitment rates during the COVID-19 pandemic. Results Baseline characteristics were broadly similar across treatment groups. Overall, 91% (52/57) patients had complex perianal fistulae;TNFi treatment had previously failed in 65% (37/57) patients. A lower proportion of patients randomized to FIL200mg than PBO discontinued the study. The proportion of patients who achieved a combined fistula response at Week24 was numerically higher in the FIL200mg than the PBO group (Figure 1a), with similar results observed for combined fistula remission (Figure 1b). Treatment-emergent severe AEs were highest in the FIL200mg group. AE rates were otherwise similar across groups. Conclusion In this phase 2 study, numerically higher fistula response and remission rates were observed after 24weeks of treatment with FIL200mg vs PBO in patients with active PFCD and a history of multiple medical treatment failures. FIL was well tolerated. Further studies of FIL for PFCD treatment are warranted.