Outcomes of two or more HLA loci mismatched unrelated donor allo-HCT: An Acute Leukemia Working Party study

The optimal allogeneic hematopoietic cell transplantation (allo-HCT) results depend on the availability of a human leukocyte antigen (HLA)-matched donor. Patients who lack an HLA-identical donor have limited donor options, specifically a haploidentical related donor, umbilical cord blood unit, or a mismatched unrelated donor (MMUD). High-resolution HLA allele level typing has improved the assessment of MMUD allo-HCT, and it has been found that increasing numbers of mismatched HLA loci correlate with a higher risk of non-relapse mortality (NRM) and graft-versus-host disease (GvHD; find out more here). However, the impact of two or more HLA mismatches on outcomes in transplanted patients with acute leukemia is unknown.

GvHD Hub Steering Committee member Arnon Nagler and colleagues performed a retrospective, multicenter study using the data set from the Acute Leukemia Working Party of the European Society for Blood and Marrow Transplantation (EBMT) to assess the outcome of patients with acute leukemia transplanted from ≥ 2 HLA allele MMUD in first or second complete remission (CR1 or CR2) who underwent first allo-HCT. Nagler et al. published their study in Bone Marrow Transplantation.¹

Study design

The inclusion criteria for this study were defined as adult patients aged ≥ 18 years with acute leukemia (acute myelogenous leukemia [AML] or acute lymphoblastic leukemia [ALL]), transplanted from ≥ 2 HLA allele MMUD (at HLA-A, -B, -C, or -DR loci between the donor and recipient), in CR1 or CR2, who underwent first allo-HCT between 2000 and 2017.

Results

A total of 145 patients with ALL and 320 patients with AML, who met the inclusion criteria, were selected for this study and followed for a median of 74.8 months (ALL) or 63.3 months (AML). The median patient age at the time of allo-HCT was higher in patients with AML compared with ALL (50.2 years and 34.8 years, respectively), and most patients had a high Karnofsky Performance Status score. Also, the majority of patients had a peripheral blood stem cell graft, myeloablative conditioning, and received in vivo T-cell depletion. Detailed baseline patient, disease, and transplant characteristics are described in Table 1.

Table 1. Baseline patient, disease, and transplant characteristics.¹

Using Cox-regression for multivariate analysis, the authors identified HLA-DR mismatch as a bad prognostic indicator, showing a remarkably higher NRM (HR, 1.67; p = 0.02), poorer leukemia-free survival (HR, 1.42; p = 0.03), overall survival (OS; HR, 1.46; p = 0.03), and higher acute (a)GVHD Grade 2–4 (HR, 1.46; p = 0.05). However, no impact was seen on relapse incidence, refined GvHD-free, relapse-free survival, or overall chronic (c)GvHD.  

Also, a higher risk of aGvHD Grade 3–4 (HR, 1.87; 95% CI, 1.08–3.22; p = 0.02) was seen in patients with more than two mismatched HLA loci while all other outcome parameters were not affected.

Finally, OS was negatively impacted by CR2 disease status (HR, 1.43; p = 0.01) and increasing patient age (per 10 years; HR, 1.17; p = 0.01).

Stem cell graft source (bone marrow vs peripheral blood) or type of conditioning regimen (MAC vs RIC) were not identified as independent factors for outcome.

The most common cause of death was disease relapse, occurring in 37% of patients with ALL and 37% of patients with AML. The second most common cause was related to infection (AML, 25% and ALL, 22%), followed by GvHD-related deaths (AML, 22% and ALL, 29%).

Results of 2-year and 5-year allo-HCT outcomes are presented in Table 2.

Table 2. Outcomes of mismatch unrelated donor hematopoietic cell transplantation.¹

Conclusion

This retrospective, multicenter study found that allo-HCT with ≤ 6 out of eight HLA allele MMUD in patients with acute leukemia led to an acceptable 2-year leukemia-free survival, and an OS which was comparable to allo-HCT from matched unrelated donors. However, mismatch at the HLA-DR locus and mismatch at more than two HLA-alleles were considered to be poor prognostic factors for the risk of NRM and aGvHD, respectively. These two factors should be considered in the donor selection process.