Haploidentical Hematopoietic Stem Cell Transplantation in Children with High Risk Hematologic Malignancies: Outcomes with Two Different Strategies for Graft-Versus-Host-Disease Prevention: Ex Vivo T-Cell Depletion and Post-Transplant Cyclophosphamide

Transplantation from a full haplotype mismatch family donor has had a striking growth in the last 20 years. A variety of different approaches are making haploidentical hematopoietic stem cell transplantation (haplo-HSCT) safer and more efficacious. These approaches include in vivo or ex vivo T-cell depletion (TCD) strategies and in vivo pharmacologic immunomodulatory strategies. Here, we compare the outcome of two cohorts of patients with high risk malignancies treated with haplo-HSCT with two different strategies for graft versus host disease (GVHD) prevention, ex vivo TCD and post-transplantation cyclophosphamide (PT-Cy). Forty patients with hematologic malignancies, median age 9 years (range, 0.5-26) underwent haplo-HSCT from 4/2005 to 4/2015. Seventeen patients were transplanted with CD3-depleted peripheral blood stem cells (PBSCs) following a reduced intensity conditioning regimen (RIC), and 23 patients received T-cell replete PBSCs followed by PT-Cy after myeloablative conditioning (MAC) (n=16) or RIC (n=7). Diagnoses included: ALL (n=17), AML (n=13), JMML (n=4), MDS (n=4), CML (n=1) and LCH (n=1); 16 patients were in CR1, 14 in CR2, 2 in CR3 and 3 had refractory diseases (RD). Diagnoses and disease status was similar in both groups. Donor engraftment occurred in 35/37 (95%) patients evaluable at day 30. Two patients (JMML) had autologous recovery in the TCD group and 3 patients (RD) died before engraftment in the PT-Cy group. Median time to neutrophils > 500/μL was 10 days (range, 9-12) in the TCD group and 15 days (range, 12-20) in the PT-Cy group (P=0.009). Platelets > 20.000/μL occurred at a median of 16 days (range, 11-23) and 20 days (range, 16-38) respectively (P=0.151). The cumulative incidence (CI) of grade II-IV and III-IV was 27% and 7% respectively in the TCD group. In the PT-Cy group the CI rate of aGVHD grade II-IV and III-IV was 45% and 5% respectively (P= NS). The CI rate of cGVHD was 9% and 53% in the TCD and PT-Cy cohorts respectively (p=0.029). Transplant related mortality (TRM) was 24% and 26% at 1 year. The CI of relapse at 2 years was 31% and 24% for the TCD and PT-Cy group respectively. With a median follow-up of surviving patients of 86 months and 17 months in each group (range, 39-128 and 8-76 months), actuarial overall survival (OS) is 47% and 48%. Causes of death were infections (n=3), sinusoidal obstructive syndrome (n=4), aGVHD (n=2) and relapse (n=9). Outcomes were acceptable and comparable in terms of engraftment, aGVHD, relapse and OS in both groups. Chronic GVHD was significantly higher in the PT-Cy approach and could be linked to the use of PBSC and the CD34 cell dose. The risk of TRM is a concern, especially for patients with advanced-stage disease. They may be not good candidates for MAC. According to our experience, results with TCD and PT-Cy are quite similar. PT-Cy strategy is simple and inexpensive. It does not seem justified the high cost and complexity of ex vivo TCD.