Efficacy and safety of human umbilical cord-derived mesenchymal stem cells versus placebo added to second-line therapy in patients with steroid-refractory acute graft-versus-host disease: a multicentre, randomized, double-blind, phase 2 trial.

backgroundFailure of systemic corticosteroid therapy is common in patients with newly diagnosed acute graft-versus-host disease (aGVHD) above grade II. Mesenchymal stem cells (MSCs) have been used as a tolerable and potentially effective second-line therapy for steroid-refractory aGVHD (SR-aGVHD); however, well-designed, prospective, controlled studies are lacking.

methodsThis multicentre, randomized, double-blind, placebo-controlled, exploratory phase 2 study enrolled patients with SR-aGVHD above grade II from 7 centres. Patients were randomized 1:1 to receive umbilical cord-derived MSCs or placebo added to one centre's choice of second-line agents (except for ruxolitinib). The agents were infused twice weekly. Patients with complete response (CR), no response (NR), or progression of disease (PD) at d28 received 8 infusions, and those with partial response (PR) received the above infusions for another 4 weeks. The per-protocol population consisted of patients who received ≥ 8 infusions. The primary endpoint was the overall response rate (ORR, CR + PR) at d28, analyzed in the per-protocol and intention-to-treat populations.

resultsSeventy-eight patients (median age 38, range 13-62) were enrolled: 40 in the MSC group and 38 in the control. Patients in the MSC group received a median of 8 doses, with a median response time of 14 days. In intention-to-treat analysis, ORR at d28 was 60% for MSC group and 50% for control group (p = 0.375). The 2-year cumulative incidence of moderate to severe cGVHD was marginally lower in the MSC group than in the control (13.8% vs. 39.8%, p = 0.067). The 2-year failure-free survival was similar between the MSC and control groups (52.5% vs. 44.4%, p = 0.43). In per-protocol analysis (n = 62), ORR at d28 was significantly greater in the MSC group than in the control group (71.9% vs. 46.7%, p = 0.043). Among patients with gut involvement, ORR at d28 was significantly greater in the MSC group than in the control (66.7% vs. 33.3%, p = 0.031). The adverse events incidences were similar between groups.

conclusionsIn this exploratory study, there was no superior ORR at d28 demonstrated in the MSC group compared with the control. However, MSCs showed a gradual treatment effect at a median of 2 weeks. Patients who completed 8 infusions may benefit from adding MSCs to one conventional second-line agent, especially those with gut involvement. MSCs was well tolerated in patients with SR-aGVHD.

trial registrationchictr.org.cn ChiCTR2000035740.