Post-Transplantation Cyclophosphamide-Based Haploidentical versus Matched Unrelated Donor Peripheral Blood Hematopoietic Stem Cell Transplantation Using Myeloablative Targeted Busulfan-Based Conditioning for Pediatric Acute Leukemia

The use of haploidentical related donors (HRDs) is a common alternative donor strategy used when matched sibling or unrelated donors are not available for hematopoietic stem cell transplantation (HSCT). However, there have been no studies comparing HRD HSCT with post-transplantation cyclophosphamide (PTCy) and matched unrelated donor (MUD) HSCT with antithymocyte globulin using similar busulfan-based myeloablative conditioning regimens in pediatric acute leukemia. Here we compared the outcomes in children and adolescents with high-risk acute leukemia who underwent HRD HSCT with PTCy (n = 35) or MUD HSCT (n = 45) after targeted busulfan-based myeloablative conditioning using intensive pharmacokinetic monitoring. The median duration of follow-up was 3.7 years in the HRD group and 4.6 years in the MUD group. No engraftment failure was observed in either group. There were no significant between-group differences in the cumulative incidence of grade II-IV acute graft-versus-host disease (GVHD) (34.3% versus 48.9%; P = .142), grade III-IV acute GVHD (2.9% versus 8.9%; P = .272), moderate to severe chronic GVHD (11.4% versus 18.3%; P = .417), relapse (25.6% versus 28.0%; P = .832), and nonrelapse mortality (0% versus 2.2%; P = .420). The 3-year severe chronic GVHD-free/relapse-free survival (GRFS), leukemia-free survival (LFS), and overall survival (OS) rates in the HRD and MUD groups were 62.9% (95% confidence interval [CI], 45.8% to 80.0%) versus 49.8% (95% CI, 34.9% to 64.7%; P = .318), 74.4% (95% CI, 58.7% to 90.1%) versus 67.5% (95% CI, 53.4% to 81.6%; P = .585), and 88.6% (95% CI, 78.0% to 99.2%) versus 83.7% (95% CI, 72.5% to 94.9%; P = .968), respectively. In a subgroup analysis of patients with acute lymphoblastic leukemia (HRD, n = 17; MUD, n = 26), the 3-year GRFS, LFS, and OS rates in the HRD and MUD groups were 49.4% (95% CI, 24.3% to 74.5%) versus 39.5% (95% CI, 19.7% to 59.3%; P = .601), 61.8% (95% CI, 37.5% to 86.1%) versus 63.6% (95% CI, 44.4% to 82.8%; P = .872), and 82.4% (95% CI, 64.4%, 100%) versus 84.2% (95% CI, 70.1% to 98.3%; P = .445), respectively. In patients with acute myelogenous leukemia (AML) (HRD, n = 16; MUD, n = 16), the 3-year GRFS, LFS, and OS rates in the HRD and MUD groups were 80.8% (95% CI, 61.2% to 100%) versus 61.9% (95% CI, 37.8% to 86.0%; P = .326), 87.1% (95% CI, 70.2% to 100%) versus 73.9% (95% CI, 51.8% to 96.0%; P = .478), and 93.8% (95% CI, 81.8% to 100%) versus 85.6% (95% CI, 67.0% to 100%; P = .628), respectively. Although the difference was not statistically significant and the number of patients was small, the promising outcomes of HRD HSCT in AML patients were encouraging. Our results demonstrate that HRD HSCT with PTCy using a targeted busulfan-based myeloablative conditioning regimen has outcomes similar to those of MUD HSCT with antithymocyte globulin. HRD HSCT with PTCy could be a feasible option for pediatric high-risk acute leukemia patients who lack an HLA-matched related or unrelated donor.