Safety and efficacy of mesenchymal stromal cell therapy for multi-drug-resistant acute and late-acute graft-versus-host disease following allogeneic hematopoietic stem cell transplantation

Document Type

Journal Article

Publication Date



Annals of hematology








Allogeneic hematopoietic stem cell transplantation; Graft-versus-host disease; Mesenchymal stromal cells


Graft versus host disease (GvHD) remains a significant risk for mortality and morbidity following allogeneic hematopoietic stem cell transplantation (HSCT). A growing literature supports successful applications of mesenchymal stromal cells (MSCs) for the treatment of steroid-refractory acute GvHD (aGvHD). However, there is limited knowledge about the effects of MSC treatment on late-acute GvHD (late aGvHD). In this article, we present our multicenter study on the safety and efficacy of MSC therapy for patients with steroid-refractory late aGvHD in comparison to those with aGvHD. The outcome measures include non-relapse mortality (NRM) and survival probability over a 2-year follow-up. The study includes a total of 76 patients with grades III-IV aGvHD (n = 46) or late aGvHD (n = 30), who had been treated with at least two lines of steroid-containing immunosuppressive therapy. Patients received weekly adipose or umbilical cord-derived MSC infusions at a dose of median 1.55 (ranging from 0.84 to 2.56) × 10/kg in the aGvHD group, and 1.64 (ranging from 0.85 to 2.58) × 10/kg in the late aGvHD group. This was an add-on treatment to ongoing conventional pharmaceutical management. In the aGvHD group, 23 patients received one or two infusions, 20 patients had 3-4, and three had ≥ 5. Likewise, in the late aGvHD group, 20 patients received one or two infusions, nine patients had 3-4, and one had ≥ 5. MSC was safe without acute or late adverse effects in 76 patients receiving over 190 infusions. In aGvHD group, 10.9% of the patients had a complete response (CR), 23.9% had a partial response (PR), and 65.2% had no response (NR). On the other hand, in the late aGvHD group, 23.3% of the patients had CR, 36.7% had PR, and the remaining 40% had NR. These findings were statistically significant (p = 0.031). Also, at the 2-year follow-up, the cumulative incidence of NRM was significantly lower in patients with late aGvHD than in patients with aGvHD at 40% (95% CI, 25-62%) versus 71% (95% CI, 59-86%), respectively (p = 0.032). In addition, the probability of survival at 2 years was significantly higher in patients with late aGvHD than in the aGvHD group at 59% (95% CI, 37-74%) versus 28% (95% CI, 13-40%), respectively (p = 0.002). To our knowledge, our study is the first to compare the safety and efficacy of MSC infusion(s) for the treatment of steroid-resistant late aGVHD and aGVHD. There were no infusion-related adverse effects in either group. The response rate to MSC therapy was significantly higher in the late aGvHD group than in the aGvHD group. In addition, at the 2-year follow-up, the survival and NRM rates were more favorable in patients with late aGVHD than in those with aGVHD. Thus, the results are encouraging and warrant further studies to optimize MSC-based treatment for late aGVHD.