Comparable Survival Outcomes of Haploidentical Stem Cell Transplantation and Unrelated Bone Marrow Transplantation

Background: In the absence of an HLA matched related donor, unrelated donor hematopoietic cell transplantation (HCT) is an alternative. Recently, HLA-haploidentical HCT using posttransplant cyclophosphamide (PTCY-haplo) has been increasingly performed. Unrelated donor peripheral blood stem cell transplantation (PBSCT) program was initiated in 2011 in Japan with rather slow increase, therefore, bone marrow is the main stem cell source for HCT from unrelated donors, while PTCY-haplo is mainly PBSCT. We compared outcomes of unrelated donor bone marrow transplant (UBMT) and PTCY-haplo.

Methods: This is a retrospective analysis of a registry data of the Japan Society for HCT and the Japanese Data Center for HCT using the Transplant Registry Unified Management Program (TRUMP). Patients with acute leukemia and myelodysplastic syndromes, aged between 16 and 69 years, who undergone their first HCT between 2012 and 2015 were included in the study. HLA-A, -B, -C, and -DRB1 allele-level 8/8 matched (n=1,470), 7/8 matched (n=859), and 6/8 matched (n=186) T-cell replete UBMT recipients, and 140 recipients of PTCY-haplo (PBSCT, n=133; BMT, n=6; PBSCT+BMT, n=1) were identified as subjects for analyses. Adjusted comparison of the groups on overall mortality was performed with the use of the Cox proportional-hazards regression model. For other outcomes with competing risks, Fine and Gray's proportional-hazards model for subdistribution of a competing risk was used. The models were used to estimate adjusted probabilities, with consideration of other significant clinical variables in the final multivariate models.

Results: The median age for 8/8 matched UBMT, PTCY-haplo, 7/8 matched, and 6/8 matched UBMT were 52(16-69), 46.5(17-68), 50(16-69), 50(16-68). According to refined disease risk index (rDRI), patients with high or very high rDRI were 32%, 54%, 34%, 34% for 8/8 matched UBMT, PTCY-haplo, 7/8 matched, and 6/8 matched UBMT. Myeloablative conditioning was used in 74%, 51%, 73%, 68% of patients for 8/8 matched UBMT, PTCY-haplo, 7/8 matched, and 6/8 matched UBMT. Median follow-up period of survivors for 8/8 matched UBMT, PTCY-haplo, 7/8 matched, and 6/8 matched UBMT were 2.79 (0.03-5.603),1.82 (015-3.89), 3.00 (0.09-5.57), 3.17 (0.27-5.58). In adjusted comparison by multivariate analyses setting 8/8 matched UBMT as the reference, PTCY-haplo showed similar overall mortality (relative risk [RR]=0.97, 95% confidence interval [CI], 0.75-1.26, p=0.823), decreased risk of non-relapse mortality (RR=0.43, 95% CI, 0.25-0.74, p=0.003), increased risk of relapse (RR=1.57, 95% CI, 1.21-2.03, p=0.001), and decreased risk of grade II to IV acute GVHD (RR=0.68, 95% CI, 0.48-0.95, p=0.023). Relative risks for 7/8 matched and 6/8 matched UBMT were 1.08 (p=0.223) and 1.27 (p=0.032) for overall mortality, 1.29 (p=0.004) and 1.73 (p<0.001) for non-relapse mortality, 0.89 (p=0.201) and 0.81 (p=0.215) for relapse, and 1.30 (p<0.01) and 1.84 (p<0.001) for grade II to IV acute GVHD. Other predictive variables identified for overall mortality were patient age older than 50 years old compared to younger at transplant, male sex, refined disease risk index of Intermediate, High, or Very High compared to Low, and HCT-CI total points of 1 or 3 or greater compared to 0. Adjusted probabilities for 8/8 matched UBMT, PTCY-haplo, 7/8 matched, and 6/8 matched UBMT at two years post-transplant were 61%, 58%, 52%, 60% for overall survival, 23%, 27%, 21%, 19% for relapse, and 20%, 6%, 24%, 33% for non-relapse mortality (Figure 1).

Conclusions: PTCY-haplo showed notably low non-relapse mortality which contributed to comparable short-term survival outcomes with 8/8 matched UBMT.

View largeDownload slide

View largeDownload slide

Close modal