Human Umbilical Cord Derived Mesenchymal Stromal Cells to Treat Steroid-Refractory Acute GvHD III/IV or Overlap Syndrome: Interim Analysis of a Multicenter Phase I/II Study

Background Treatment of patients developing steroid-refractory acute GvHD (SR-aGvHD) after allogeneic hematopoietic cell transplantation (HCT) remains an unmet clinical need and a major therapeutic challenge. Mesenchymal stromal cells (MSC) induce immunosuppression and reduce the inflammatory status initiated by aGvHD. We performed an interim analysis of a multicenter, prospective, non-randomized, phase I/II study, planned to test the safety and activity of human umbilical cord (UC), third-party MSC given sequentially after Pentatostatin for the treatment of SR-aGvHD grade III-IV as well as the overlap syndrome (EUDRACT n. 2012-000582-21), NCT02032446). Patients and Methods Third-party UC MSC were produced under GMP conditions as previously described (Capelli C et al Cytotherapy, 2011, 13, 786-801). The treatment schedule was based on the sequential administration of Pentostatin, given iv at dose of 1 mg/m^2 for 3 consecutive days, followed by 3 MSC infusions given at weekly intervals. Between October 2013 and May 2018 (range 4.6 years) we enrolled 27 allogeneic HCT adult recipients (22 male) with SR-aGvHD (n 25) or overlap syndrome (n 2), median age 47 years (range 19-62) transplanted from HLA-identical sibling donors (6; 22%), haploidentical donors (2; 7%), unrelated donors (18; 67%) or cord blood (1; 4%). Target organ involvement included skin in 11 cases (44%), gastrointestinal tract (GI) in 23 (92%) and liver in 12 (48%). Nine patients had grade III (33%), 16 patients grade IV (60%) and 2 patients had severe overlap syndrome (7%). Most of patients exhibited multiple organ involvement (n 16, 64%). Patients started MSC therapy at median of 22 days (10-2101) after the onset of aGvHD or overlap syndrome. Results The infusions of UC MSC-infusion were always well tolerated without any immediate or late toxic event. Response to MSC therapy, by day 30 after the final MSC dose, was evaluable in 16 of 27 patients: 7 achieved CR (26%), 6 PR (22%) and 3 NR (11%). Ten patients died before reaching the evaluation time of response: 8 patients for aGvHD and 2 for haematological disease. One patient was lost to follow up after 9 days from the last UC-MSC infusion. Remarkably, the subgroup of patients with single organ involvement (8 GI, 1 liver) were all evaluable showing 6 patients in overall response (67%, 4 RC and 2 PR). No response was observed in 2 cases of severe overlap syndrome. Patients achieving CR had improved longer term OS vs patients in PR/NR (80% vs 50%