Introduction:Graft-versus-host disease (GvHD) remains to be a factor associated with significant morbidity and mortality after allogeneic hematopoietic stem cell transplantation (HSCT). Bortezomib was shown in vitro to induce selective depletion of alloreactive T-lymphocytes, decrease the production of Th1 cytokines (Blanco et al. 2006), and to suppress the maturation and cytokine production by dendritic cells (Naujokat et al. 2007). Bortezomib was successfully used to prevent and treat GVHD in different settings (Al-Homsi et al. 2015; Koreth et al. 2015). As GvHD-prophylaxis agent Bortezomib was implemented since 2014 in aim to improve the results in a cohort of pediatric patients with acute leukemia who underwent HSCT from HLA-matched unrelated and haploidentical family donors after TCRαβ/CD19-depletion. Here we present the results of a retrospective comparison of two GvHD-prophylaxis regimens.Patients and methods:Between May 2012 and July 2016 eighty-one transplantation from 40 HLA-matched unrelated and 41 haploidentical donors were performed for children (56 boys and 25 girls with median age 9 years, range 0,6-23 years) with acute lymphoblastic (n=31) and acute myeloblastic (n=50) leukemia in complete remission. TCRαβ/CD19-depletion of HSC with CliniMACS technology was implemented in all cases. For all patients it was first allogeneic HSCT.The majority (93%) of the patients received Treosulfan-based condition regimen. Remaining 7% of patients received TBI-based conditioning regimen.Patients were divided retrospectively in two groups according to GvHD prophylaxis regimens. "Regimen 1" (n=35), which was used in 2012-2013 yrs.: horse ATG 50 mg/kg and post-transplant tacrolimus with short course of methotrexate. "Regimen 2" (n=46) started in 2014: rabbit ATG 5 mg/kg, rituximab 200mg/m2 and peri-transplant bortezomib 1,3 mg/m2 on days -5, -2, +2 and +5. The median dose of CD34+ cells in the transplant was 9 x106/kg (range 7-17), TCRαβ - 21x103/kg (range 1-305).Results:Groups differed significantly in regards to diagnosis: 57% of patients in "Regimen 1" had ALL, while most of the patients (78%) in "Regimen 2" were with AML (P= 0,002).Cumulative incidence of neutrophil and platelet engraftment at 30 days was 98% and did not differ between the groups. Median time to neutrophil and platelet engraftment was 14 days, (range, 9-33 and 9-25 days, respectively). Neutrophil engraftment was significantly faster among patients with "Regimen 2", 13 days vs. 16 days for patients with "Regimen 1" and CI of engraftment at day 30 after HSCT 98% (95% CI: 98%) vs. 94% (95% CI: 87-100, P<0,01).Overall cumulative incidence of acute GvHD II-IV grade was 23% (95% CI: 16-35); grade III-IV - 5% (95% CI: 2-13) and chronic GvHD - 18% (95% CI: 11-20). Cumulative incidence of acute GvHD II-IV was significantly lower within the group with "Regimen 2": 15% (95% CI: 7-30) vs. 34% (95% CI: 22-54), P=0,05. Amid patients with "Regimen2" there was one case of grade IV acute GvHD, most of the patients with grade II to IV developed visceral damage involving lower gut. "Regimen 2" was also more effective in prevention of chronic GvHD: CI at 1 year after HSCT was 7% vs. 31%, P=0,005. Only one patient with "Regimen 2" had extensive form of chronic GvHD.Median time of follow-up for survivors was 2 years (range, 0,3 - 4). Cumulative incidence of relapse at 2 years also differed between "Regimen 1" and "Regimen 2" groups, 31% (95% CI: 19-51) vs. 21% (95% CI: 11-39), respectively, though without statistical significance. TRM was 10% (95% CI: 5-20), without significant statistical difference between leukemia type, donor type or GvHD-prophylaxis regimens. EFS (event=death or relapse) at 2 years was 64% (95%CI: 53-75), OS - 69% (95%CI: 58-80). Statistically there was no significant difference in event-free or overall survival probabilities between leukemia type, donor type or GvHD-prophylaxis regimensConclusion:In our retrospective single-center study we revealed that rATG, rituximab and bortezomib improve the control of acute and chronic GVHD in recipients of TCRαβ- depleted grafts in comparison to hATG, tacrolimus and methotrexate. DisclosuresNo relevant conflicts of interest to declare.