Background: Complex perianal fistulas are a serious complication in patients with Crohn’s disease (CD). Darvadstrocel (DVS), a suspension of expanded adult allogeneic adipose-derived mesenchymal stem cells, is approved in Europe and Japan for the treatment of complex Crohn’s perianal fistulas (CPF). The global ADMIRE-CD II phase 3 randomized double-blind placebo-controlled study evaluated the efficacy and safety of DVS for the treatment of complex CPF.

Methods: Patients aged 18–75 years with clinically controlled, inactive or mildly active luminal CD and complex CPF (≤ 2 internal openings [IOs] and ≤ 3 external openings [EOs]) who had an inadequate, or a loss of response to immunosuppressive agents or biologics were included. Patients were randomized 1:1 to receive a local injection of DVS as a single dose (120 x 10⁶ cells/24 mL) or placebo. In all patients, fistula preconditioning included vigorous curettage and seton placement 2–3 weeks before treatment, and seton removal, a further curettage and closure of IOs immediately before treatment. The primary endpoint was combined remission (closure of all treated EOs that were draining at baseline, despite gentle finger compression, and absence of collections > 2 cm confirmed by magnetic resonance imaging [MRI]) at 24 weeks. Secondary endpoints included combined remission at 52 weeks, clinical remission (closure of all treated EOs without MRI confirmation) at 24 and 52 weeks, and time to clinical remission at 24 weeks. Safety was monitored up to 52 weeks.

Results: From 19 October 2017 to 26 July 2023, 568 patients received DVS (n = 283) or placebo (n = 285) with fistula preconditioning; 56.3% of patients enrolled in Europe and Israel, and 43.7% in North America. Mean (SD) age, sex and race were similar in the DVS and placebo arms (38.4 [11.9] vs 37.7 [10.8] years; 42.8% vs 45.6% female; 85.9% vs 89.1% White). Combined remission rates at 24 weeks did not statistically differ between treatments (48.8% DVS vs 46.3% placebo) and there were no differences in secondary endpoints (Table 1, Figure 1). Based on health authority guidelines, post hoc analyses of patients randomized before COVID-19 (11 March 2020; n = 141 DVS, n = 143 placebo) were performed: combined remission rates at 24 weeks were 46.8% (DVS) and 38.5% (placebo). The safety profile for DVS was consistent with prior studies with no new safety signals (Table 1).

Conclusions: The efficacy outcomes assessed did not statistically differ between DVS and placebo, and the placebo response rate (with fistula preconditioning) was higher than expected. Post hoc analyses revealed lower placebo response rates in patients randomized before COVID-19 (similar to the pivotal ADMIRE-CD I study) than the overall placebo arm. DVS was well tolerated.

Pradeep Nazarey

Appearances