Finding an HLA-matched related (MRD) or unrelated donor (UD) is not always possible, so different studies have been carried out to establish alternative donors such as the haploidentical donor.
The objective of this study was to compare the results of allo-SCT in patients with acute myeloid leukemia (AML) according to the type of donor: MRD, UD or haploidentical donor. We also analyzed other characteristics of donors: Age, sex, cytomegalovirus (CMV) status and ABO incompatibility.
Retrospective unicentric study. We analyzed 43 patients with AML undergoing myeloablative allo-SCT between January 2013 and January 2018. The median age was 42 (31-53) years. The cytogenetic risk groups were established according to the European Leukemia Net 2017 (ELN 2017) classification. The characteristics of the patients and donors are listed in Table 1. The prophylaxis for graft-versus-host disease (GVHD) was performed with ciclosporin in all cases together with methotrexate in the case of MRD, thymoglobulin in UD and cyclophosphamide in haploidentical donors. To estimate overall survival (OS) and progression free survival (PFS), we used the Kaplan-Meier statistical analysis. To correlate quantitative variables with qualitative multivariate variables, we used ANOVA analysis. To correlate clinical parameters we used the Chi-square test.
Analyzing by subgroups the types of donors HLAi (MRD, UD and haploidentical), we did not find statistically significant differences in the average of days of neutrophil graft (> 0.5 × 10^3/μL). On the other hand, we found differences in the average of days of platelet graft (> 50 × 10^3/μL), (12 ± 3 days in MRD, vs 15 ± 3 days in UD, vs. 20 ± 5 days in haploidentical donors, p <0.0001). We did not find differences between type of donor and prevalence of bacterial or fungal infections during the first 6 months after transplantation or in CMV reactivation prevalence. We found differences between type of donor and GVHD prevalence, it was less prevalent in haploidentical donors (64% vs 8%, p = 0.001). In addition, in patients with a haploidentical donor, a tendency to lower PFS was obtained compared to non-haploidentical donors (35 vs 54 months, HR 2.98, 0.86-10.3, p = 0.08) (graph 1). However, we did not find differences in OS based on the type of donor (graph 2). We also did not find differences in OS, PFS, the average of days of neutrophil or platelet graft, GVHD, bacterial or fungal infections during the first 6 months depending on other donor characteristics such as age, sex, CMV status or ABO incompatibility.
Haploidentical donor is an option for allo-SCT, presenting SG results similar to MRD. The use of cyclophosphamide in haploidentical transplantation produces a significant immunosuppression in patients, which could justify the results obtained, with a lower GVHD rate, with lower PFS too, but without affecting OS.