Hematology, Transfusion and Cell Therapy (Oct 2024)

PTCY-BASED HAPLOIDENTICAL TRANSPLANTATION, WITH REDUCED-DOSE MMF, 30 MG/KG, AND A CALCINEURIN INHIBITOR, COMPARED WITH ATG-BASED, 6 MG/KG, UNRELATED DONOR TRANSPLANTATION FOR HEMATOLOGICAL MALIGNANCIES

  • LJ Arcuri,
  • CO Ribeiro,
  • IO Fernandes-Junior,
  • SC Maradei,
  • SP Lermontov,
  • D Lerner

Journal volume & issue
Vol. 46
pp. S1024 – S1025

Abstract

Read online

Introduction: The EBMT recommends an ATG (Thymoglobulin) dose of 4.5‒6.0 mg/kg for an unrelated (URD) hematopoietic cell transplant (HCT). The PTCy-based Baltimore protocol for haploidentical (Haplo) HCT included MMF 45 mg/kg starting at D+5, however MMF 30 mg/kg starting at D0 has also been reported. In our institution, the first 6 Haplo received 45 mg/kg MMF, with a 50% incidence of CMV disease, prompting us to reduce MMF to 30 mg/kg. Objective: To compare 6.0 mg/kg ATG-based URD versus PTCy-based with MMF 30 mg/kg Haplo HCT. Methods: This single-center study was conducted in a public hospital in Rio de Janeiro, Brazil. We included all patients who underwent Haplo or URD HCT for acute leukemia or myelodysplasia between 2016 and 2024. URD received 6 mg/kg ATG with a Calcineurin Inhibitor (CNI) and an antimetabolite. Haplo received PTCy combined with a CNI and approximately, due to rounding, MMF 30 mg/kg. All follow-up times were longer than 30 days. Survival and cumulative incidence curves were carried out with Kaplan-Meier and Gray methods and compared with logrank and Gray tests. Propensity Score (PS) weighted-Cox models were performed to control for relevant variables. The study was approved by the Ethics Committee and registered at ClinicalTrials (NCT02759822). Results: With a median follow-up of 13- and 28-months, 74 Haplo and 83 URD were included, respectively. Except for conditioning intensity (reduced-intensity conditioning in 30% of Haplo and 6% of URD) and graft (peripheral blood in 51% of Haplo and 35% of URD), all other variables were relatively well-balanced. In the URD group, 82% were HLA 8/8 matched, while 18% were 7/8 matched. 2-y OS, for Haplo and URD, respectively, were 51% and 70% (p = 0.10); 2-y RFS 51% and 64% (p = 0.23); GRFS 22% and 47% (p = 0.01); 2-y relapse 17% and 18% (p = 0.62); 2-y NRM 32% and 17% (p = 0.08); 2-y cGVHD 39% and 35% (p = 0.52), 2-y moderate/severe cGVHD 22% and 15% (p = 0.49); 6-month II‒IV aGVHD 50% and 49% (p = 0.72); and 6-month III‒IV aGVHD 15% and 23% (p = 0.51). In the PS-weighted model, only GRFS was different: HR = 1.70 for Haplo vs. URD, p = 0.01. In other PS-weighted models, nor OS, HR = 1.50, p = 0.17; PFS, HR = 1.40, p = 0.24; relapse, HR = 1.00, p = 0.97; II‒IV aGVHD, HR = 1.10, p = 0.63; III‒IV aGVHD, HR = 1.30, p = 0.56; cGVHD, HR = 1.40, p = 0.24; nor moderate and severe cGVHD, HR = 1.70, p = 0.23 were statistically different, although, numerically, most outcomes favored the URD group. Discussion: Our results suggest it might not be safe to systematically offer Haplo transplants for patients with suitable URD. PS-weighted GRFS was statistically improved with URD, compared with Haplo. Overall, our Haplo results were comparable with those already reported, although, in most comparisons of Haplo versus URD, ATG was not given to all URD patients. Importantly, GVHD control was not hampered by MMF reduction from 45 mg/kg to 30 mg/kg in the Haplo arm, even in a highly miscegenated country like Brazil. Time to transplant is usually shorter in Haplo, and therefore we cannot rule out survival bias favoring the URD group. Although not randomized, it was a relatively large sample that included patients with only hematologic malignancies.