Post-Transplant Cyclophosphamide Versus Antithymocyte Globulin in Patients with Acute Myeloid Leukemia Undergoing Allogeneic Stem Cell Transplantation from One Antigen HLA-Mismatched Donors: A Retrospective Analysis from the Acute Leukemia Working Party of the EBMT

Introduction. Several alternative donor sources are currently available for patients who lack an HLA-matched related or unrelated donor, including haploidentical, cord blood and one antigen HLA-mismatched donors (9/10 mMUD). Use of post-transplant (allo-HSCT) cyclophosphamide (PTCY) as graft-versus-host disease (GVHD) prophylaxis allowed outcome improvements in the haploidentical setting. Its use has also been reported as a safe and feasible option in 9/10 mMUD transplants. However, to date, the main strategy used as GVHD prophylaxis in 9/10 mMUD allo-HSCT is in vivo T-cell depletion with antithymocite globulin (ATG). Data comparing these two different anti GVHD prophylaxis strategies in 9/10 mMUD allo-HSCT are limited.

Methods. We compared PTCY versus ATG as GVHD prophylaxis in patients undergoing 9/10 mMUD allo-HSCT for which high-resolution HLA-allele typing was available in the ALWP/EBMT data registry. Included were adult patients (age>18 years) undergoing their first allo-HSCT for acute myeloid leukemia (AML) during the period 2007-2017. All disease status were allowed. Propensity score matching was performed to reduce and eliminate confounding effects. Each patient receiving PTCY was matched with two patients receiving ATG using the nearest neighbor or exact matching. Variables included in the propensity score model were: disease status at time of allo-HSCT, conditioning regimen, age, secondary AML, female donor to male recipient, source of stem cells, patient and donor CMV serology status.

Results. Globally, 93 patients receiving PTCY were identified and matched with 179 patients receiving ATG. Secondary AML was reported in 20% and 18% of patients in ATG and PTCY groups, respectively. Most patients were in first complete remission at time of allo-HSCT (55% and 56% in PTCY and ATG group, respectively), while nearly 29% of patients in both groups underwent allo-HSCT with active disease. Ciclosporine (csA) and mycophenolate mofetil (MMF) were the systemic immunosuppressive agents more frequently associated to either PTCY (42%) or ATG (49%). Other well represented associated immunosuppressive agents were tacrolimus and MMF in ATG group (20%), followed by 14% of patients receiving csA alone with ATG. In PTCY group, 39% of patients received csA and methotrexate as associated immunosuppressive agents. Conditioning regimen was myeloablative in 50% of patients in both groups. Peripheral blood was the preferred stem cell source (91% in both groups). Among the variables not included in the propensity score model, some differences were observed between the two groups. Median follow-up was longer in the ATG group (27.4 versus 14.2 months, p<0.01). Gender was more frequently male in PTCY group (60% versus 45%, p<0.02). Median year of allo-HSCT was 2014 and 2015 in ATG and PTCY groups, respectively (p<0.01). Similar engraftment rates were observed for both groups (95% and 96% in PTCY and ATG groups, respectively). At 2 years, leukemia-free survival (LFS) was higher in patients receiving PTCY (55% vs 35%, p<0.05), severe (grade III-IV) acute GVHD was lower (9% vs 19%, p<0.04) and GRFS higher (37% vs 21%, p<0.02) as compared to patients receiving ATG, while no differences were observed in relapse incidence (RI)( 29% vs 37%, p=0.31), overall survival (OS, 56% vs 38%, p=0.06) and in non-relapse mortality (NRM) (16% vs 29%, p=0.06). Main causes of death were similarly distributed in both groups, with infectious complications being more frequently represented (25% and 22% in the PTCY and ATG groups, respectively). Grade II-IV acute GVHD (30% vs 32%, p=0.39) and chronic GVHD (39% vs 36%, p=0.35) were also similar in the two groups.

Conclusions. In patients undergoing 9/10 mMUD allo-HSCT, use of PTCY as GVHD prophylaxis is a safe and feasible option, representing a possible valid and superior alternative to ATG. Use of PTCY, indeed, may ensure a lower incidence of severe acute GVHD and higher LFS and GRFS as compared to ATG. These registry based results may serve the basis for a prospective randomized trial comparing PTCY to ATG as anti GVHD prophylaxis in 9/10 mMUD allo-HSCT.