Long-term survival after unrelated donor marrow transplantation for aplastic anaemia after optimized conditioning regimen: a retrospective multicentre cohort study

Abstract

Almost all acquired severe aplastic anaemia is immune mediated and characterised by hypocellular bone marrow and ≥2 affected haematopoietic lineages. The optimal preparartive regimen for unrelated donor transplantation remains to be established. We aimed to study long-term outcomes after unrelated donor transplantation for severe aplastic anaemia with de-escalation of cyclophosphamide (Cy) dose in steps of 50mg/kg (150, 100, 50, 0mg/kg) in combination with total body irradiation (TBI) 2Gy, anti-thymocyte globulin (ATG) and fludarabine. Ninety-six patients with severe aplastic anaemia aged ≤65 years with adequate organ function enrolled on a trial of human leukocyte antigen (HLA)-matched or 1 HLA-locus mismatched unrelated donor marrow transplantation conducted between 02/2006 and 12/2013 in the United States (NCT00326417). Exclusion criteria were Karnofsky performance status of less than 60, clonal cytogenetic abnormalities and inherited marrow failure syndormes. The primary outcome was day-100 engraftment (achievement of absolute neutrophil recovery to at least 0.5×109/L without subsequent decline) and day-100 survival. The trial determined the lowest effective Cydose as 50mg/kg (n=38) for day-100 engraftment and survival. Cy dose 100mg/kg (n=41) was also acceptable. Accrual to Cy doses 150mg/kg (n=15) and 0mg/kg (n=3) was terminated early for toxicities. The current study is an extended follow up of patients enrolled on the trial (NCT00326477) and includes 76 of 96 patients alive ≥1 year after transplantation. There were 20 deaths in the first year after transplantation (Cy 0mg/kg [n=2], Cy 50mg/kg [n=1], Cy 100mg/kg [n=10], Cy 150mg/kg [n=7]). Patients were followed prospectively from transplantation and data reported using standardized data collection forms until death, loss to follow up or last contact through November 2023. The incidence of graft failure was calculated using the cumulative incidence estimator and the probability of survival using the Kaplan-Meier estimator. The median follow up of the cohort is 8.02 (IQR) 5.16-10.12) years. With Cy 50mg/kg, there was one graft failure and five deaths ≥1 year after transplantation. With Cy 100mg/kg there was only one late death and no graft failure. The 8-year probabilities of survival were 85.0% (95% CI 67.3-93.5) and 75.6% (95% CI 59.4-86.1) after Cy50mg/kg and 100mg/kg, respectively, P=0.31. With Cy 0mg/kg and 150mg/kg, there were no graft failures or death ≥1 year after transplantation. Regardless of Cy dose 12 of 15 patients aged ≥50 years died. Cy 50mg/kg or 100mg/kg with TBI 2Gy, ATG and fludarabine are effective conditioning regimens for unrelated donor marrow transplants for aplastic anaemia. Identification of an optimized transplantation approach for patients aged ≥50 years is needed. US National Institutes of Health.