The impact of specific loci mismatches in 9/10 HLA transplants

Stem cell transplantation (SCT) remains one of the best methods of consolidation in patients with acute myeloid leukemia (AML). Patients without a matched sibling donor or matched unrelated donor generally receive a 9/10 matched transplant. In a study published in Bone Marrow Transplantation, J. Loke and colleagues investigated whether a mismatch of alleles at the HLA-A, -B, -C, -DRB1 (DR), or -DQB1 (DQ) locus was better than another.¹

Study design

In this retrospective study, 937 patients with AML, who were treated with reduced intensity conditioning allogeneic (allo) SCT, were analyzed. Primary endpoints were non relapse mortality (NRM), relapse incidence (RI), acute graft-versus-host disease (aGvHD), and chronic GvHD (cGvHD). Baseline patient characteristics can be found in Table 1 and are fairly homogenous, with no significant differences observed. High-resolution genotyping of alleles was done at the HLA-A, -B, -C, -DR, and -DQ loci. As can be seen from Table 1, mismatch occurred most frequently at A and C, and least at DQ. In total, 80% had de novo AML. Most patients (> 72%) received in vivo T-cell depletion, and the majority (62%) were given antithymocyte globulin, while only 19.7% had alemtuzumab.

Table 1. Patient characteristics¹

Key findings

The median follow-up period was 24 months. Multivariate analysis showed that RI, leukemia-free survival (LFS), and overall survival (OS) were significantly impacted by age and high-risk cytogenetics. Survival outcomes are shown in Table 2. The cytomegalovirus (CMV) status of the recipient also had a negative impact on OS, specifically if they were seropositive (HR, 1.3; 95% CI, 1.05−1.16; p = 0.016). The risk of aGvHD was reduced in patients undergoing transplant more recently (2010−2015) (HR, 0.71; 95% CI, 0.54−0.92; p = 0.011) but NRM was not affected.

Table 2. Survival outcomes at 2 years and GvHD incidence¹

Mismatch univariate analysis was performed, and it was discovered that HLA-DQ mismatch was demonstrated to be favorable over other loci in relation to NRM. In contrast, increased NRM was seen with mismatch at the HLA-A locus, which appeared to be associated with higher incidence of aGvHD.

In a further Cox proportional-hazards regression analysis, HLA-DQ was used as the reference for comparison. No loci were preferential regarding LFS, OS, RI, and GvHD relapse free survival. For NRM on the other hand, HLA-A mismatch was significantly associated with increased risk (HR, 1.75; 95% CI, 1.14−2.67; p = 0.01), which may be linked to an increased risk of aGvHD (HR, 1.79; 95% CI, 1.24−2.6; p = 0.002).

The effect of CMV status was investigated and donor CMV status had no impact on survival outcomes. In contrast, if the recipient was seropositive for CMV, a higher rate of aGvHD was noted, leading to a reduced OS (p = 0.01). This latter association was also confirmed by multivariate analysis. NRM was higher when patients were seropositive (HR, 1.67; 95% CI, 1.21−2.30) as was the incidence of aGvHD (HR, 1.33; 95% CI, 1−1.75; p = 0.046). However, in the same analysis, CMV seropositivity in the donor decreased the risk of aGvHD (HR, 0.78; 95% CI, 0.61–1.01; p = 0.06).

The risk of both aGvHD (HR, 0.67; 95% CI, 0.5–0.9; p = 0.009) and cGvHD (HR, 0.62; 95% CI, 0.45–0.84; p = 0.002) was lessened when patients underwent in vivo TCD, however, relapse risk was unaffected.

Conclusion

Overall, there were no significant differences in OS and LFS between mismatches at different loci, although HLA-A mismatch was associated with increased NRM. These results show that 9/10 HLA matched donors are suitable for use as donors in patients undergoing reduced intensity conditioning  allo-SCT without a sibling or matched donor.

Irrespective of donor status, recipients who were seropositive for CMV showed an increased risk of aGvHD and NRM while LFS and OS were decreased. This should prompt screening of patients and increased CMV prophylaxis and treatment.