Transplant using a KIR B donor offers relapse protection to patients with AML undergoing reduced-intensity conditioning

Natural killer (NK) cells play an important role in the outcome of allogeneic hematopoietic stem cell transplantation (allo-HSCT) by promoting engraftment, preventing acute graft-versus-host disease (GvHD), supporting immune reconstitution, and reducing the risk of leukemic relapse.^1,2 Their activity is influenced by signaling via activating and inhibitory receptors.

Killer-cell immunoglobulin-like receptors (KIRs) upon binding to HLA class I epitopes can impede NK cell’s ability to recognize and eliminate malignant target cells. Therefore, a donor’s KIR gene haplotype influences the outcome of allo-HSCT for acute myeloid leukemia (AML). A transplant from donors with a KIR B haplotype has been reported to protect against relapse after myeloablative unrelated donor (MUD) allo-HSCT for AML.^3,4

Previous reports suggest that conditioning regimen (reduced-intensity conditioning, RIC vs myeloablative conditioning, MAC) can affect NK cell reconstitution, that correlates with clinical outcomes.⁵ Although MAC MUD allo-HSCT has been shown to improve relapse protection and increase disease-free survival (DFS) in recipients of  donors with KIR B haplotype, the impact of RIC in this setting has not been previously examined.⁵

A stronger protection from relapse for KIR B haplotype donors, has been demonstrated for recipients with the C1 epitope, and with a HLA-C-mismatched transplant, including HLA-C1/C2 mismatch.⁵

Daniel Weisdorf, University and Minnesota, US, and colleagues evaluated whether the impact of MUD KIR genotype on transplant outcome for patients with AML has changed due to improvements in transplant practice in recent years.⁶

Methods

Patient and donor demographics, transplant approach, and outcome data were collected between 2012–2016 from the prospective KIR DS trial⁷ and the contemporaneous cohort, through the Center for International Blood and Marrow Transplant Research. KIR genotypes for the contemporaneous cohort were collected retrospectively through the typing project of the National Marrow Donor Program.⁸ Donor KIR genotypes were used to assign KIR AA or Bx haplotypes. KIR haplotype is defined by the number and composition of activating and inhibitory genes. The KIR A haplotype is characterized by the presence of seven KIR genes, including six inhibitory and one activating. The KIR B haplotype is characterized by the presence of a varied number of KIR genes, with at least one activating KIR and two inhibitory genes. The impact of demographic and donor KIR genotype on relapse incidence, non-relapse mortality (NRM), DFS, and overall survival (OS) was evaluated in patients receiving MAC vs RIC.

 Results⁶

• The prospective cohort included 243 transplantations, with 36 months median follow-up of survivors
• The contemporaneous retrospective cohort included 2,419 transplantations, with 44 months median follow-up of survivors
• Overall, patients had similar characteristics across both cohorts, except for a smaller proportion of CMV seropositivity and lower number of HLA < 8/8 allele-matched donors. In the prospective study cohort compared with the contemporaneous group cohort. The prospective cohort had more patients ATG/alemtuzumab naïve, and increased level of co-morbidity, disease status and donor age.

Clinical outcomes of allo-HSCT

• Majority of patients engrafted (98–99%) and had 11–13% of transplant-related NRM at six months
• In both cohorts, conditioning (RIC vs MAC) failed to show a significant impact on the incidence of relapse
  • Estimated 5-year survival rates were not significantly different
    • RIC recipients – 39% and 44% in the contemporaneous and prospective trial cohorts, respectively
    • MAC recipients – 49% and 45% in the contemporaneous and prospective trial cohorts, respectively
  • 5-year DFS rates were not significantly different
    • RIC recipients – 35% and 40% in the contemporaneous and prospective trial cohorts, respectively
    • MAC recipients – 46% and 38% in the contemporaneous and prospective trial cohorts, respectively

Analysis of the RIC cohort

Previous data showed a protective role of donor KIR B haplotype after myeloablative unrelated donor (MUD) allo-HCT for AML, with recipients having lower risk of relapsed. Transplant using KIR AB and BB vs AA haplotype donor, significantly reduced the risk of relapse (HR = 0.77; 95% CI, 0.62–0.97; p = 0.026) and increased DFS (HR = 0.84; 95% CI, 0.72–0.99; p = 0.038), especially in patients with the fully HLA-matched donors (Table 1).

Table 1. Significant factors in multivariate analysis for risk of relapse and independently significant factors (p < 0.05) DFS in patients after reduced intensity conditioning. Full findings can be observed in the original manuscript⁶

CI, confidence interval; DFS, disease-free survival; HR, hazard ratio; KIR, killer-cell immunoglobulin-like receptor

Factor n HR 95% CI p value Relapse

Donor KIR  haplotype

AA

Bx

 

334

730

 

1

0.77

 

0.62–0.97

0.026

 

0.026

Cytogenetics 0.0025

Disease status

Early

Advanced

 

83

79

 

1

3.97

 

1.88–8.36

0.001

 

0.0003

DFS

HLA matched alleles

7/8

8/8

 

145

917

 

1

0.78

 

0.61–0.99

0.042

 

0.042

Cytogenetics 0.0012

Disease status

Early

Intermediate

Advanced

 

836

218

8

 

1

1.18

3.01

 

0.98–1.43

1.22–7.4

0.022

 

0.085

0.017

Recipient CMV status 0.035

Impact of recipient HLA epitopes on the outcome in the RIC cohort

There was a lower risk of relapse and a longer DFS with donor KIR Bx  haplotypes in HLA-C1/x vs C2/C2 recipients (HR, 0.76; 95% CI, 0.61–0.97; p = 0.024 and HR, 0.85; 95% CI, 0.72–0.99; p = 0.04, respectively). Those patients also had greater OS (HR, 0.82; 95% CI, 0.68–0.98; p = 0.029).

Conclusion

The data demonstrated that transplants from KIR B haplotype donors where recipients expressed the C1 epitope of HLA-C offered a significantly reduced risk of leukemia relapse and improved DFS after RIC allo-HSCT. Conditioning using fully myeloablative regimens did not offer such benefits. The authors believe that the findings are important as RIC is commonly used in the current allo-HCT practice. Therefore, selection of KIR B donors for MUD allo-HSCT could offer significant relapse protection in patients with AML undergoing allo-HSCT.