Inhibitory KIR-Ligand Interactions and Relapse Protection Following HLA Matched Allogeneic HCT for AML

Elizabeth Krieger, Rehan Qayyum, Amir Toor
doi: https://doi.org/10.1101/2020.07.10.20149047

Abstract

Killer immunoglobulin-like receptor (KIR) and KIR-ligand (KIRL) interactions play an important role in natural killer cell mediated graft versus leukemia effect (GVL) after hematopoietic stem cell transplant (HCT) for AML. Mathematically accounting for KIR-KIRL interactions may identify donors with optimal NK cell mediated alloreactivity and GVL. In this retrospective study of 2359 donor recipient pairs (DRP) who underwent unrelated donor (URD) HCT for AML, KIR-KIRL interaction scores were determined. Relapse risk was significantly reduced in donor-recipient pairs (DRP) with higher inhibitory KIR-KIRL interaction and missing KIRL (mKIR) scores, with HR=0.86 (P=0.01) & HR=0.84 (P=0.02) respectively. This effect was not observed with activating KIR-KIRL interactions. The inhibitory KIR-KIRL (iKIR) interaction score components were summed to give an inhibitory-missing ligand (IM-KIR) score, which if it was 5 as opposed to <5, was also associated with a lower relapse risk, SHR 0.8 (P=0.004). Acute and chronic graft vs. host disease (GVHD), survival, GRFS and relapse free survival were not significantly different. However, TRM was increased among those with IM-KIR=5, HR, 1.32 (P=0.01). Among those with HLA matched DRP, anti-thymocyte globulin recipients with IM-KIR=5, had a lower relapse rate HR, 0.61 (p=0.001), however TRM was increased in these patients with a HR, 1.49 (p=0.034). This study demonstrates that unrelated DRPs with high inhibitory KIR content scores confer relapse protection, albeit with increased TRM. Clinical trials utilizing donors with a higher iKIR content in conjunction with novel strategies to reduce TRM should be considered for URD HCT recipients with AML to optimize clinical outcomes.

Highlights Inhibitory and missing KIR-KIRL interactions, both individually and collectively are predictive of relapse following URD HCT for AML.

Amongst 8/8 HLA matched donor recipient pair relapse reduction was observed in those undergoing in vivo T cell depletion with ATG.

TRM is increased in patients with greater magnitude of inhibitory and missing KIR-KIRL interactions.

Introduction

Acute myeloid leukemia (AML) is the most common form of adult acute leukemia and indication for hematopoietic cell transplantation (HCT) from human leukocyte antigen (HLA) matched unrelated donors (URD)1,2. Advances in donor selection, conditioning regimens and supportive measures have improved overall survival for patients with AML, however, relapse rates after HCT have remained stagnant1,3–7,8. The curative potential of HCT is predicated largely on graft vs leukemia (GVL) effect of the transplant9. Natural killer cells (NK) have been shown to provide cell-mediated alloreactivity, potentially by targeting non-antigen binding domains of the HLA molecules through germline-encoded receptors, such as, killer immunoglobulin-like receptors (KIR)10,11. These KIR have either inhibitory or activating function and interact with killer immunoglobulin-like receptor ligands (KIRL) on the target cells, specifically HLA molecules12. The effect of KIR-KIRL interactions in a HLA-matched URD HCT setting have long been debated with conflicting results 13–22. These inconsistencies may stem from the non-quantitative, descriptive methodology used to examine KIR-KIRL ligand interactions. These include a description of KIR-A and KIR-B haplotypes based on activating KIR gene content, and the presence of certain activating and inhibitory KIR genes and their ligands. Therefore, KIR genotyping, while commercially available, has not yet been established as a standard of care in unrelated donor selection as of yet, despite extensive laboratory evidence of the efficacy of NK cells in mediating anti-leukemia activity in vitro23.

This disconnect between known in vitro effective GVL mediation and apparent lack thereof in vivo, may be overcome by accounting for the complexity encountered in NK cell alloreactivity which is determined by multiple KIR and KIRL interacting simultaneously. A novel mathematical approach to quantifying KIR-KIRL interactions has been published24, utilizing a system of matrix, vector-operator equations to account for all possible donor-recipient KIR-KIRL interactions. This analysis highlighted the previously unknown importance of donor inhibitory KIR (iKIR) interactions, which complement the known missing KIRL interactions (mKIR) in mediating GVL.

In this study, the effect of known KIR-KIRL interactions on HCT outcomes in patients with AML is retrospectively examined in a large cohort of recipients of HLA matched and HLA mismatched URD to validate the findings reported earlier. Given the relevance of NK cells in mediating GVL effects following allogeneic HCT, the hypothesis that HCT for AML performed in donor-recipient pairs (DRP) with a higher donor inhibitory KIR-KIRL interaction would result in improved clinical outcomes, primarily by a reduction in relapse rates is examined.

Methods

Study design

The analysis presented here was performed on deidentified donor and recipient demographic and clinical outcomes data, provided by the Center for International Blood and Marrow Transplant Research (CIBMTR, Milwaukee, WI). In brief, data were available for 2,408 DRPs transplanted for AML between 2010 and 2016 at 113 centers across the United States. All donors were unrelated, and transplants facilitated via the National Marrow Donor Program. Patient variables examined included age, sex, Karnofsky performance status, HLA match grade (8/8 or 7/8, including HLA-A, -B, -C and -DRB1), KIR and HLA ligand status (A3, A11, Bw4 or Bw6, C1 or C2). KIR genotyping of all known KIR genes was available for all donors and was performed as previously described25. HLA typing was reported at high-resolution and KIR typing as presence or absence. Disease related characteristics included cytogenetic risk (SWOG criteria) and disease state at the time of transplant early (CR1) or late (CR 2-3). Cytogenetics were categorized as good, intermediate, poor; as a large number of observations were missing cytogenetics information (N=1494), a fourth category of ‘missing’ cytogenetics was included. Transplant-related variables included conditioning intensity, graft source, the use of in vivo T cell depletion (ATG or Alemtuzumab), and GVHD prophylaxis regimens utilized. Donor related characteristics included donor age, and KIR gene presence or absence.

KIR KIRL interaction scores

KIR-KIRL interaction values were assigned as previously described24,26. Known KIR with established KIRL were utilized and are outlined in supplementary table 1. Essentially, the model assumes that NK cell interaction with its target cells may be mathematically described as a series of matrix equations, where the NK cell with its KIR molecules constitutes a ‘vector’, and the target with its KIRL molecules, an ‘operator’. The operator (target) modifies the vector (NK cell) upon interacting with it, either inhibiting or activating it. Each individual KIR-KIRL interaction may then be described as follows; if an inhibitory KIR (iKIR; represented by −1) on the NK cell encounters a ligand on its target (represented by 1), this will give the NK cell an inhibitory signal (overall effect −1). Mathematically this interaction which may then be represented by the equation, (−1) × (1) = −1, assuming a constitutively active basal state for NK cells. Conversely, if there is no ligand for an inhibitory KIR, i.e., missing KIRL (mKIRL; represented by −1), the interaction score will be (−1) × (−1) = +1 because of the abrogation of the inhibitory signal and persistent NK cell activation. Finally, activating KIR (aKIR; represented as +1) interacting with its ligands will be scored, (1) × (1) = +1, when the ligand is present, and (1) × (0) = 0, when the ligand is absent, since no activating signal is given. Each of these different scores constitutes a distinct component of the total KIR effect on individual NK cells expressing them, the cumulative effect determining the eventual outcome of NK cell – target cell interaction. The absolute value of each DRP’s KIR component scores, iKIR, mKIR and aKIR, were then summed to give the respective scores (Supplementary Figure 1, Supplementary Table 2). The absolute value of each KIR component score was utilized to allow comparison of the relative magnitude of each component regardless of the direction of NK cell effect. This implies that iKIR score corresponds to the number of iKIR in the donor encountering their corresponding ligand in the recipient, while mKIR score is the number of iKIR in the donor missing their corresponding ligand in the recipient. The iKIR and mKIR scores were then summed up to calculate the composite inhibitory-missing ligand (IM)-KIR score, iKIR + mKIR = IM KIR. A logic-based algorithm (Supplementary Table 3) was utilized to calculate the KIR-KIRL interaction scores for the entire cohort in accordance with these equations and examples of iKIR, mKIR, aKIR and IM KIR Scores are given in Supplementary table 2.

View this table:
Supplementary Table 1.

KIR and corresponding KIRL population frequency in the study cohort.

View this table:
Supplementary Table 2:

Examples of KIR scoring in 2 potential DRP.

View this table:
Supplementary Table 3.

Algorithm to generate KIR-KIRL interaction component scores and the IM-KIR score.

Supplementary Figure 1.
Supplementary Figure 1.

NK cell (blue) with inhibitory KIR (red chevron) and activating KIR (green chevron) interacting with targets which either express, or do not express cognate ligands (red and green triangles). The KIR input collectively determines NK cell function, i.e., NK cell mediated alloreactivity.

Outcome Definition

Relapse and relapse-free survival (RFS) were defined per CIBMTR criteria. Transplant related mortality (TRM) was defined as death in absence of relapse. Overall survival (OS) was defined as the time from HCT to last follow up or death from any cause. Chronic GVHD (cGVHD) was defined using NIH consensus criteria. GVHD free – relapse free survival was defined as freedom from relapse and cGVHD27.

Statistical Methods

Characteristics between the patients who relapsed and who did not were compared using Pearson’s chi-square test or Student’s t-test. Kaplan-Meier survival curves by dichotomous IM-KIR scores were compared using the log-rank test. To evaluate the influence of confounding variables on the effect of NK cell mediated alloreactivity as determined by the KIR component scores, a multivariate Cox proportional hazards model was constructed including biologically relevant, clinical and demographic data available. Competing risk analysis treated TRM as a competing risk for relapse; analysis for TRM included relapse as a competing event; and for cGVHD occurrence, relapse and TRM before cGVHD diagnosis were treated analogously. Cox proportional hazard models were also generated using acute GVHD (grades 2-4 and 3-4), overall survival, and cGVHD-free, relapse-free survival as outcomes. These models were generated without and with adjustment for potential confounders based on prior literature and biological plausibility. To examine if the inclusion of missing cytogenetic data observations as a separate category might have affected the study results, sensitivity analyses were conducted using only observations without missing data. We performed additional exploratory analysis to understand the relationship between IM-KIR score and clinical outcomes in clinically important subgroups; these differences between the subgroups were examined using an interaction term in the models. All analyses were performed in Stata 15.0 (College Station, Texas) and a P-value <0.05 was considered significant. Since this study was designed to validate the findings reported previously, i.e., iKIR & mKIR individually and cumulatively impact clinical outcomes, multiple comparisons were not adjusted for.31

Results

Distribution of KIR-KIRL interaction scores in HLA matched URD SCT recipients

After excluding patients who failed to engraft neutrophils (N=30) or had missing data on relapse or time to relapse (N=12), the study cohort was comprised of 2365 donor recipient pairs (DRP) who underwent unrelated donor HCT for early or intermediate AML (Table 1). This cohort included adults aged 20-83 (mean 53) years; the majority (85%) of DRPs were high-resolution 8/8 HLA-matched for HLA-A, HLA-B, HLA-C, and HLA-DRB1. All patients received T cell replete grafts; 42% (n=996) received in vivo T cell depletion, 937 (94%) with ATG and 59 (6%) with Alemtuzumab. The majority 86% received a graft of mobilized peripheral blood stem cells (PBSC), 59% received myeloablative conditioning. KIR and KIRL gene frequencies within this population are given in Supplementary Table 1. This cohort was primarily of Caucasian descent (89%).

View this table:
Table 1:

Study population characteristics; total population and by IM-KIR scores.

Absolute values of the KIR-KIRL interaction scores varied across the population regardless of the HLA match grade (8/8 vs 7/8) and different ethnic groups of donors and patients. Absolute values of the KIR-KIRL interaction component scores were; iKIR median 3 (range: 1 to 5); mKIR median 2 (range: 0 to 4); aKIR median 0 (range: 0 to 4). The median IM-KIR score was 4, (range: 2 to 5) (Figure 1).

Figure 1.
Figure 1.

Population frequencies of KIR-KIRL interaction scores for inhibitory KIR (iKIR), missing KIRL (mKIR), activating KIR (aKIR) and IM-KIR scores. (N=2316).

Inhibitory KIR scores impact relapse risk following HLA matched URD SCT

The effect of all 3 KIR-KIRL interaction component scores (absolute values of iKIR, mKIR and aKIR) on relapse was evaluated as a continuous variable. Higher iKIR and mKIR scores, were significantly associated with a reduced risk of AML relapse, with sub-hazard ratios (SHR) of 0.86 and 0.85 respectively (Figure 2). However, aKIR did not significantly impact relapse risk for AML. These effects were observed independent of several confounding variables included in the multivariable analysis, specifically, recipient and donor ages, conditioning regimen intensity and in vivo T cell depletion, which if unfavorable (older donor age, RIC, and ATG use) all increased the risk for relapse. No significant effect of the KIR-KIRL interactions was observed on other clinical outcomes examined (acute GVHD, overall and relapse-free survival or TRM). The likelihood of cGVHD was higher in those with a higher iKIR score (SHR 1.09), albeit not statistically significantly.

Figure 2.
Figure 2.

Unadjusted and adjusted sub-distribution hazard and hazard ratios from competing risk and Cox proportional hazard models for the effect of iKIR, mKIR and aKIR on clinical outcomes following HCT for AML. Sensitivity analysis performed to account for missing values of confounding variables.

A dichotomous IM-KIR score delineates relapse risk

Given their independent predictive value for relapse following URD HCT, the cumulative effect of iKIR and mKIR component scores was then examined with the IM-KIR score. IM KIR score was dichotomized for analysis into those DRP with an IM KIR score of 5 (IM=5), and those with an IM-KIR scores of 2, 3 or 4 (IM<5). The cut-off value was chosen to have relatively balanced groups as few DRPs had IM KIR scores of 2 or 3 (Figure 1). As compared with DRPs in the IM<5 group, patients in the IM=5 group had a 20% lower risk of relapse (SHR 0.80; Figures 3 & 4). Patients in the IM=5 group remained at a lower risk of relapse, (SHR 0.80), when adjusted for other relevant demographic variables (Table 2). As observed with the two component scores (iKIR & mKIR), the relapse protection was independent of older recipient and donor ages (>49 years), ATG use and RIC usage, which were all associated with increased relapse risk.

View this table:
Table 2:

Unadjusted and adjusted results of the competing risk and Cox proportional hazard models for the effect of IM-KIR score on clinical outcomes for URD HCT for AML. SHR (95% CI); P-values reported. During the 1,965,798 person-day follow up, 747 patients relapsed, and 1,175 patients died.

Figure 3.
Figure 3.

Unadjusted and adjusted sub-distribution hazard and hazard ratios from competing risk and Cox proportional hazard models for the effect of DIM Scores on clinical outcomes following HCT for AML. Sensitivity analysis performed to account for missing values of confounding variables.

Figure 4.
Figure 4.

Unadjusted cumulative incidence curves by IM-KIR scores for relapse, treatment-related mortality, and chronic graft versus host disease, as well as failure curves by IM-KIR scores for overall survival, relapse-free survival, and cGVHD-free, relapse-free survival.

Patients in the IM=5 group also had a higher risk of developing cGVHD (SHR 1.13) in unadjusted analysis (Figure 3 & 4), but, as with individual KIR-KIRL interaction components this did not remain statistically significant on multivariate analysis (Table 2). Acute GVHD was also not impacted by the IM-KIR score. TRM risk was increased in the IM=5 patients, but this difference was not independent of increased recipient age, poor KPS, HLA mismatch and cyclosporine-based GVHD prophylaxis which, increased the risk for TRM (Table 2). Despite the strong relapse protection effect, there was not a survival or relapse-free survival benefit observed within the two IM-KIR score categories.

IM KIR score and relapse protection in recipients of HLA matched allografts

To determine the effect of variation in IM-KIR score and outcomes in distinct groups of patients, interaction analyses were performed. First, the interaction between IM-KIR score and degree of HLA matching was examined in the entire cohort. In those transplanted with an 8/8 HLA matched donor, who were also IM=5, there was a significant reduction in relapse risk (SHR 0.76, p=0.001) as compared to the reference group (IM<5), however this was not so among the 7/8 HLA match and IM=5 donor recipients (Supplementary Figure 2). The interaction analysis did not disclose a significant relationship between the degree of HLA mismatch and relapse reduction in the IM=5 patients (interaction P =0.12). As the degree of HLA match did not modify IM-KIR effect on relapse protection, and in general HLA mismatch was associated with poor survival (Table 2), further subset analysis was limited to 8/8 HLA-matched (n=1958) DRP.

Supplementary Figure 2.
Supplementary Figure 2.

Adjusted sub-distribution hazard and hazard ratios for relapse from competing risk and Cox proportional hazard models for the effect of IM=5 on clinical outcomes following HCT for AML in subgroups.

Among those transplanted with an 8/8 HLA matched donor, 730 patients received ATG for in vivo T cell depletion: of those, the patients who received grafts from IM<5 donors, 38% had relapsed within 3 years, while among those transplanted with IM=5 donors, only 26% relapsed (SHR 0.61) (Table 3, Figure 5). Notably, the IM-KIR score for the donor modified the effect of the use of ATG on relapse risk (interaction p=0.049), i.e., the DRP with an IM=5 donor, when administered ATG, did not experience the increased relapse risk seen in IM<5 DRP who received ATG (Table 3, Figure 5). On the other hand, among those patients who did not receive any in vivo T cell depletion the 3-year relapse rates were not significantly different in those transplanted with IM=5 (27%) vs. IM<5 (29%) donors. Relapse reduction was also seen in IM=5 DRP among most subgroups of patients examined (Table 3).

View this table:
Table 3:

Subgroup analysis of 8/8 HLA matched DRP, with alemtuzumab patients removed. Unadjusted and adjusted results of the competing risk, Cox proportional hazard models and interaction analysis for the effect of IM-KIR score on clinical outcomes for URD HCT for AML. SHR (95% CI); P-values reported.

Figure 5.
Figure 5.

Unadjusted cumulative incidence curves by IM-KIR scores and in vivo T cell depletion with ATG depicting relapse, treatment-related mortality, and chronic graft versus host disease.

IM-KIR score influence on transplant related mortality and GVHD

Despite the relapse risk reduction seen in IM=5 donors, there was an absence of survival, RFS and GRFS benefit. This is related to the increased TRM observed in the 8/8 HLA matched, IM=5 DRP as compared to IM<5 donors (SHR 1.32; Table 3). This effect appeared to be most prominent among those who received ATG and had IM=5 donors (SHR 1.49; Table 3; Figure 5). The TRM increase may be contributed to by a higher, albeit non-significant likelihood of cGVHD developing in those who did not receive ATG and had an IM=5 donor as opposed to IM<5 (SHR 1.15; Table 3, Figure 5).

Discussion

In this study a novel, logic-based mathematical-model examining the KIR-KIRL interactions is reported as a surrogate measure of NK cell alloreactivity in patients with AML undergoing URD HCT. In this large cohort of HLA matched and single antigen mismatched DRPs, a higher iKIR and mKIR score, as well as combined IM-KIR score predicted a lower risk of relapse in recipients transplanted from a donor with a higher iKIR content. This relapse risk reduction was most evident in HLA matched donor-recipient pairs (DRP) undergoing transplantation with in vivo T cell depletion using ATG. Despite this relapse risk reduction, no survival advantage was observed, due to a higher risk of TRM observed in the DRP with the highest inhibitory KIR content.

The mathematical approach to examining KIR-KIRL interactions utilized here is similar to the model system utilized in studying the interaction of T cell receptors with minor histocompatibility antigen-HLA complexes28. The vector (NK cell) – operator (target cell) interaction model allows for all the known interactions to be quantified in a sound mathematical framework, and the clinical associations of this cumulative interaction studied. Somewhat contrary to the current dogma, this analytic methodology has identified a strong association of iKIR donor-recipient interactions with AML relapse protection. It is known that in vitro models have shown that NK cells with an increased number of iKIRs that interact with their requisite HLA are more responsive when they encounter a target cell 29,30. This suggests that cells reconstituting in a post HCT setting with the potential for a larger number of iKIR-HLA interactions may potentially have higher graft vs leukemia (GVL) activity. The results reported here support that counterintuitive notion, which may be understood using the analogy of an electrical capacitor. In this instance the iKIR interactions of a NK cell may be considered analogous to the dielectric of a capacitor. Like a capacitor which stores electric charge, NK cells may be considered to possess alloreactive killing potential. Capacitors with a stronger dielectric (insulator) capacity can store a larger amount of charge, likewise NK cells with increased iKIR interactions, in the early post-transplant setting may be inhibited by the vast majority of cells with appropriate iKIRL on their surfaces and retain a larger ‘killing capacity’ for when they encounter a potential target. Downregulation of HLA, class I and II molecules on AML blasts is well described immune evasion mechanism for escape from T cell alloreactivity31–33. However, the primed iKIR-rich-NK cells upon encountering such blasts may provide more robust effector function.

Further, in the converse situation NK cells with a greater aKIR content which encounter aKIRL will experience constant activating input (weaker dielectric and insulation), and potentially be activated in a nonspecific fashion, eventually being tuned down and rendered anergic to prevent auto-reactivity34. This quantitative perspective of NK cell interactions with normal vs. leukemic cells would suggest that iKIR interactions may increase the specificity of the NK cell alloreactivity, making it more likely to effect a GVL response as observed here (Figure 6). Indeed, the results reported here suggest that the greater the iKIR content of the donor, the greater the benefit in terms of relapse protection, regardless of the number of iKIR ligands present.

Figure 6.
Figure 6.

NK cell alloreactivity model. Numerous activating aKIR-KIRL interactions (green receptors & ligands) between aKIR expressing NK cells (Blue) and normal (light grey) as well as malignant cells (dark grey), imply that the NK cells cannot specifically target leukemic blasts, which may lead to non-specific and thus, ineffective NK cell mediated alloreactivity. With iKIR-KIRL interactions (red receptors & ligands) where activation only occurs upon interaction with iKIRL low/negative cells, more specific and targeted GVL activity may be observed.

The relapse benefit of the IM=5 was most evident and significant in HLA matched allograft recipients of in vivo T cell depleted (TCD) grafts where ATG was used. This was in contrast with the overall higher relapse risk in ATG recipients in the cohort as a whole, and in IM<5 donors in particular. Further, the IM=5 and IM<5 patients who did not get ATG had no significant difference in relapse risk, albeit at a higher risk of developing cGVHD. One may reason that, higher NK cell alloreactivity of the greater iKIR and mKIR interactions (IM=5) attenuated the higher relapse risk observed when ATG was given and cGVHD mediated alloreactivity and relapse protection therefrom neutralized. Similar relapse protection was observed in recipients of RIC conditioning, suggesting an NK cell mediated alloreactive mechanism of relapse reduction.

Despite the lower relapse risk no survival benefit of IM=5 donors observed in this study, rather, a higher TRM was seen. The cumulative incidence of TRM in recipients given ATG with IM=5 donors is higher than in those getting ATG with an IM<5 donor, who had the lowest TRM risk, despite an equal rate of cGVHD with the ATG and IM=5 donors. This suggests that the additional TRM risk may be related to infections in donors with higher iKIR content, rather than GVHD. From an evolutionary standpoint, NK cells protect the host from viral infections35–38. Many viruses, in particular the Herpesviridae, such as cytomegalovirus and Epstein-Barr virus have developed elaborate mechanisms for evading NK cell and CD8+ T cell mediated lysis of infected cells. These include deployment of decoy HLA class I receptors, such as UL 18 in CMV which would increase the inhibitory input to the NK cells. IM=5 as opposed to IM<5 donor derived NK cells may be more susceptible to such viral suppression of NK cell effect of combating viral infection, contributing to TRM. Interestingly, amongst HLA matched patients those who received PBSC, and had donors with IM=5, a lower relapse rate was observed, without increases in TRM, perhaps because of the higher T cell dose delivered with these grafts.

HLA matching has long been the gold standard for URD selection, with an ever increasing level of precision in matching achieved through high resolution typing39. However beyond that there has been little which distinguishes equivalently HLA matched donors, until recently when large data sets established donor age to be of definite prognostic value in URD transplant outcomes40,41. Even though it is well established that NK cells reconstitute early following HCT42, and the effect of KIR-KIRL interactions on NK cell mediated alloreactivity has been well established16,22,43, KIR genotyping has not been successfully applied to donor selection44. Greater activating KIR content or KIR B haplotypes have been reported to have a favorable influence on clinical outcomes45,46. However, in this large cohort, iKIR and mKIRL scores, had a protective role vis a vis relapse. The data presented here give a mathematical framework for incorporating KIR genotyping in URD selection for future clinical trials.

This model is simple, and its underlying mathematics provide the ability to grow and adapt as new discoveries are made regarding NK cell alloreactivity and KIR interactions in a post-transplant setting, particularly for the aKIR for which the ligands are not known yet. However, it is important to note that at this moment it considers all KIR HLA interactions uniformly. It is known that some KIR, for example, KIR 2DL2 binds HLA C with a higher affinity then KIR 2DL3, this means that the model may underestimate or overestimate the effect of specific interactions47,48. The different relative contributions of certain KIR-KIRL interactions will need to be examined in future studies to overcome this limitation but can easily be accommodated in the mathematical framework of the model utilized. Effects of NK cell receptors like NKG2 family of receptors and their ligands on the target cells may also be incorporated in this vector – operator model system. Further, these data also raise an important mechanistic questions, why should iKIR and mKIR have the same magnitude of effect? Aspects of NK cell biology not accounted for in this model may contribute to this; for example, NK cell education, or, the time dependent expression of KIR following HCT, as well as any effect KIRL-zygosity might have on these interactions.

In conclusion, a strong relapse protection effect was observed in patients transplanted with donors with an IM-KIR score of 5. This challenges the notion that KIR are irrelevant to donor selection, and raises the question that those donors with the highest IM KIR scores in an appropriate transplant setting, may be considered as optimal donors for HCT recipients with AML to allow for increased NK cell mediated alloreactivity. Future clinical trials evaluating donor selection for URD HCT should include this measure to evaluate its value prospectively in uniformly treated patient cohorts, with adequate GVHD and antiviral prophylaxis to mitigate TRM.

Data Availability

The data are available through the CIBMTR.

Conflict-of-Interest Disclosure

The authors have no relevant conflicts of interest to disclose.

Acknowledgements

The authors gratefully acknowledge Stephen Spellman and Tao Wang, PhD for their help with data organization and for critical review of the manuscript. The analysis presented here was performed on data obtained from the Center for International Blood and Marrow Transplantation Research (CIBMTR, Milwaukee, WI). The CIBMTR is supported primarily by Public Health Service grant/cooperative agreement U24CA076518 with the National Cancer Institute (NCI), the National Heart, Lung and Blood Institute (NHLBI) and the National Institute of Allergy and Infectious Diseases (NIAID); grant/cooperative agreement U24HL138660 with NHLBI and NCI; grant U24CA233032 from the NCI; grants OT3HL147741, R21HL140314 and U01HL128568 from the NHLBI; contract HHSH250201700006C with Health Resources and Services Administration (HRSA); grants N00014-18-1-2888, N00014-17-1-2850 and N00014-20-1-2705 from the Office of Naval Research; subaward from prime contract award SC1MC31881-01-00 with HRSA; subawards from prime grant awards R01HL131731 and R01HL126589 from NHLBI; subaward from prime grant award 5R01CA2151343 from NIH Cancer Institute. P01CA111412, R01CA152108, R01CA218285;R01CA231141, R01HL126589, R01HL129472, R01HL131731 and U01AI126612 from the NIH. The views expressed in this article do not reflect the official policy or position of the National Institute of Health, the Department of the Navy, the Department of Defense, Health Resources and Services Administration (HRSA) or any other agency of the U.S. Government. AT was supported by research funding from the NIH-NCI Cancer Center Support Grant (P30-CA016059; PI: Gordon Ginder, MD). Institutional Review Board at the Virginia Commonwealth University deemed that since data were completely deidentified, this analysis did not constitute human subjects research and according to Section 45 CFR 46.102(l) of the HHS Regulations for the Protection of Human Subjects did not require IRB scrutiny.

EK: Designed study, developed KIR-KIRL scoring algorithm, analyzed data, wrote paper.

RQ: Designed study, developed KIR-KIRL scoring algorithm, performed statistical analysis, wrote paper.

AT: Designed study, developed KIR-KIRL scoring algorithm, analyzed data, wrote paper.

References

  1. 1.↵
    O’Donnell MR, Tallman MS, Abboud CN, et al. Acute myeloid leukemia, version 3.2017: Clinical practice guidelines in oncology. JNCCN J Natl Compr Cancer Netw. 2017;15(7):926–957. doi:10.6004/jnccn.2017.0116
    OpenUrlAbstract/FREE Full Text
  2. 2.↵
    Döhner H, Estey E, Grimwade D, et al. Diagnosis and management of AML in adults: 2017 ELN recommendations from an international expert panel. Blood. 2017;129(4):424–447. doi:10.1182/blood-2016-08-733196
    OpenUrlAbstract/FREE Full Text
  3. 3.↵
    Schlenk RF. Post-remission therapy for acute myeloid leukemia. Haematologica. 2014;99(11):1663–1670. doi:10.3324/haematol.2014.114611
    OpenUrlAbstract/FREE Full Text
  4. 4.
    Lee SJ, Klein J, Haagenson M, et al. High-resolution donor-recipient HLA matching contributes to the success of unrelated donor marrow transplantation. Blood. 2007;110(13):4576–4583. doi:10.1182/blood-2007-06-097386
    OpenUrlAbstract/FREE Full Text
  5. 5.
    Harvey J, Green A, Cornish J, et al. Improved survival in matched unrelated donor transplant for childhood ALL since the introduction of high-resolution matching at HLA class i and II. Bone Marrow Transplant. 2012;47(10):1294–1300. doi:10.1038/bmt.2012.8
    OpenUrlCrossRefPubMed
  6. 6.
    Fleischhauer K, Shaw BE, Gooley T, et al. Effect of T-cell-epitope matching at HLA-DPB1 in recipients of unrelated-donor haemopoietic-cell transplantation: A retrospective study. Lancet Oncol. 2012;13(4):366–374. doi:10.1016/S1470-2045(12)70004-9
    OpenUrlCrossRefPubMedWeb of Science
  7. 7.↵
    Al-Homsi AS, Roy TS, Cole K, Feng Y, Duffner U. Post-transplant high-dose cyclophosphamide for the prevention of graft-versus-host disease. Biol Blood Marrow Transplant. 2015;21(4):604–611. doi:10.1016/j.bbmt.2014.08.014
    OpenUrlCrossRefPubMed
  8. 8.↵
    Passweg JR, Labopin M, Cornelissen J, et al. Conditioning intensity in middle-aged patients with AML in first CR: No advantage for myeloablative regimens irrespective of the risk group-an observational analysis by the Acute Leukemia Working Party of the EBMT. Bone Marrow Transplant. 2015;50(8):1063–1068. doi:10.1038/bmt.2015.121
    OpenUrlCrossRefPubMed
  9. 9.↵
    Bleakley M, Riddell SR, Cancer FH. Molecules and Mechanisums of the Graft-Versus-Lleukemia. Nat Rev. 2004;4(May):1–10. doi:10.1038/nrc1365
    OpenUrlCrossRef
  10. 10.↵
    Gill S, Olson JA, Negrin RS. Natural Killer Cells in Allogeneic Transplantation: Effect on Engraftment, Graftversus-Tumor, and Graft-versus-Host Responses. Biol Blood Marrow Transplant. 2009;15(7):765–776. doi:10.1016/j.bbmt.2009.01.019
    OpenUrlCrossRefPubMedWeb of Science
  11. 11.↵
    Mehta RS, Rezvani K. Immune reconstitution post allogeneic transplant and the impact of immune recovery on the risk of infection. Virulence. 2016;7(8):901–916. doi:10.1080/21505594.2016.1208866
    OpenUrlCrossRef
  12. 12.↵
    Cooley S, Parham P, Miller JS. Review Series STRATEGIES TO IMPROVE GRAFT-VERSUS-LEUKEMIA EFFECTS Strategies to Activate NK Cells to Prevent Relapse and Induce Remission Following Hematopoietic Stem Cell Transplantation.; 2018. www.bloodjournal.org.
  13. 13.↵
    Stringaris K, Marin D, Barrett AJ, et al. KIR gene haplotype: An independent predictor of clinical outcome in MDS patients. Blood. 2016;128(24):2819–2823. doi:10.1182/blood-2016-05-713099
    OpenUrlAbstract/FREE Full Text
  14. 14.
    Escudero A, Martínez-Romera I, Fernández L, et al. Donor KIR Genotype Impacts on Clinical Outcome after T Cell–Depleted HLA Matched Related Allogeneic Transplantation for High-Risk Pediatric Leukemia Patients. Biol Blood Marrow Transplant. 2018;000:1–8. doi:10.1016/j.bbmt.2018.08.009
    OpenUrlCrossRef
  15. 15.
    Blazar BR, Noreen H, Weisdorf DJ, et al. Negative effect of KIR alloreactivity in recipients of umbilical cord blood transplant depends on transplantation conditioning intensity. Blood. 2009;113(22):5628–5634. doi:10.1182/blood-2008-12-197467
    OpenUrlAbstract/FREE Full Text
  16. 16.↵
    Davies SM, Ruggieri L, DeFor T, et al. Evaluation of KIR ligand incompatibility in mismatched unrelated donor hematopoietic transplants. Blood. 2002;100(10):3825–3827. doi:10.1182/blood-2002-04-1197
    OpenUrlAbstract/FREE Full Text
  17. 17.
    Arima N, Kanda J, Tanaka J, et al. Homozygous HLA-C1 is Associated with Reduced Risk of Relapse after HLA-Matched Transplantation in Patients with Myeloid Leukemia. Biol Blood Marrow Transplant. 2018;24(4):717–725. doi:10.1016/j.bbmt.2017.11.029
    OpenUrlCrossRef
  18. 18.
    Giebel S, Nowak I, Dziaczkowska J, et al. Activating killer immunoglobulin-like receptor incompatibilities enhance graft-versus-host disease and affect survival after allogeneic hematopoietic stem cell transplantation. Eur J Haematol. 2009;83(4):343–356. doi:10.1111/j.1600-0609.2009.01280.x
    OpenUrlCrossRefPubMed
  19. 19.
    Venstrom JM, Pittari G, Gooley TA, et al. HLA-C-Dependent Prevention of Leukemia Relapse by Donor Activating KIR2DS1 A bs t r ac t. N Engl J Med. 2012;367:805–821. doi:10.1056/NEJMoa1200503
    OpenUrlCrossRefPubMedWeb of Science
  20. 20.
    Miller JS, Cooley S, Parham P, et al. Missing KIR ligands are associated with less relapse and increased graft-versus-host disease (GVHD) following unrelated donor allogeneic HCT. Blood. 2007;109(11):5058–5061. doi:10.1182/blood-2007-01-065383
    OpenUrlAbstract/FREE Full Text
  21. 21.
    Hsu KC, Keever-Taylor CA, Wilton A, et al. Improved outcome in HLA-identical sibling hematopoietic stem-cell transplantation for acute myelogenous leukemia predicted by KIR and HLA genotypes. Blood. 2005. doi:10.1182/blood-2004-12-4825
    OpenUrlAbstract/FREE Full Text
  22. 22.↵
    Ruggeri L, Capanni M, Urbani E, et al. Effectiveness of Donor Natural Killer Cell Alloreactivity in Mismatched Hematopoietic. Vol 295.; 2002. http://www.jstor.org/stable/3076290%5Cnhttp://www.jstor.org/stable/.
  23. 23.↵
    Vivier E, Santo J Di, Moretta A. Current Topics in Microbiology and Immunology Natural Killer Cells.; 2016. http://www.springer.com/series/82.
  24. 24.↵
    Krieger E, Sabo R, Moezzi S, et al. Killer Immunoglobulin-like Receptor-Ligand Interactions Predict Clinical Outcomes Following Unrelated Donor Transplants. Biol Blood Marrow Transplant. October 2019. doi:10.1016/j.bbmt.2019.10.016
    OpenUrlCrossRef
  25. 25.↵
    Weisdorf D, Cooley S, Wang T, et al. KIR Donor Selection: Feasibility in Identifying better Donors. Biol Blood Marrow Transplant. 2019;25(1):e28–e32. doi:10.1016/j.bbmt.2018.08.022
    OpenUrlCrossRef
  26. 26.↵
    Cooley S, Parham P, Miller JS. Strategies to activate NK cells to prevent relapse and induce remission following hematopoietic stem cell transplantation. Blood. 2018;131(10):1053–1062. doi:10.1182/blood-2017-08-752170
    OpenUrlAbstract/FREE Full Text
  27. 27.↵
    Park SS, Jeon YW, Min GJ, et al. Graft-versus-Host Disease–Free, Relapse-Free Survival after Allogeneic Stem Cell Transplantation for Myelodysplastic Syndrome. Biol Blood Marrow Transplant. 2019;25(1):63–72. doi:10.1016/j.bbmt.2018.08.004
    OpenUrlCrossRef
  28. 28.↵
    Abdul Razzaq B, Scalora A, Koparde VN, et al. Dynamical System Modeling to Simulate Donor T Cell Response to Whole Exome Sequencing-Derived Recipient Peptides Demonstrates Different Alloreactivity Potential in HLA-Matched and -Mismatched Donor-Recipient Pairs. Biol Blood Marrow Transplant. 2016;22(5):850–861. doi:10.1016/j.bbmt.2015.11.1103
    OpenUrlCrossRef
  29. 29.↵
    Johansson S. The strength of inhibitory input during education quantitatively Human Immune system variation View project Symposium: A quest towards curing childhood cancer by pinpointing better molecular biomarkers View project Petter Brodin Karolinska Institutet. 2014. doi:10.1182/blood-2008-05-156836
    OpenUrlAbstract/FREE Full Text
  30. 30.↵
    Yu J, Heller G, Chewning J, Kim S, Yokoyama WM, Hsu KC. Hierarchy of the Human Natural Killer Cell Response Is Determined by Class and Quantity of Inhibitory Receptors for Self-HLA-B and HLA-C Ligands. J Immunol. 2007;179(9):5977–5989. doi:10.4049/jimmunol.179.9.5977
    OpenUrlAbstract/FREE Full Text
  31. 31.↵
    Verheyden S, Ferrone S, Mulder A, et al. Role of the inhibitory KIR ligand HLA-Bw4 and HLA-C expression levels in the recognition of leukemic cells by Natural Killer cells. Cancer Immunol Immunother. 2009;58(6):855–865. doi:10.1007/s00262-008-0601-7
    OpenUrlCrossRefPubMed
  32. 32.
    Jan M, Leventhal MJ, Morgan EA, et al. Recurrent genetic HLA loss in AML relapsed after matched unrelated allogeneic hematopoietic cell transplantation. Blood Adv. 2019;3(14):2199–2204. doi:10.1182/bloodadvances.2019000445
    OpenUrlAbstract/FREE Full Text
  33. 33.↵
    Stölzel F, Hackmann K, Kuithan F, et al. Clonal evolution including partial loss of human leukocyte antigen genes favoring extramedullary acute myeloid leukemia relapse after matched related allogeneic hematopoietic stem cell transplantation. Transplantation. 2012;93(7):744–749. doi:10.1097/TP.0b013e3182481113
    OpenUrlCrossRefPubMedWeb of Science
  34. 34.↵
    Pittari G, Liu X-R, Selvakumar A, et al. NK Cell Tolerance of Self-Specific Activating Receptor KIR2DS1 in Individuals with Cognate HLA-C2 Ligand. J Immunol. 2013;190(9):4650–4660. doi:10.4049/jimmunol.1202120
    OpenUrlAbstract/FREE Full Text
  35. 35.↵
    De Pelsmaeker S, Romero N, Vitale M, Favoreel HW. Herpesvirus Evasion of Natural Killer Cells. J Virol. 2018;92(11). doi:10.1128/jvi.02105-17
    OpenUrlCrossRef
  36. 36.
    Jackson SE, Redeker A, Arens R, et al. CMV immune evasion and manipulation of the immune system with aging. GeroScience. 2017;39(3):273–291. doi:10.1007/s11357-017-9986-6
    OpenUrlCrossRef
  37. 37.
    Wilkinson GWG, Tomasec P, Stanton RJ, et al. Modulation of natural killer cells by human cytomegalovirus. J Clin Virol. 2008;41(3):206–212. doi:10.1016/j.jcv.2007.10.027
    OpenUrlCrossRefPubMedWeb of Science
  38. 38.↵
    Mancini M, Vidal SM. Mechanisms of Natural Killer Cell Evasion Through Viral Adaptation. Annu Rev Immunol. 2020;38(1):511–539. doi:10.1146/annurev-immunol-082619-124440
    OpenUrlCrossRef
  39. 39.↵
    Dehn J, Spellman S, Hurley CK, et al. Special Report Selection of Unrelated Donors and Cord Blood Units for Hematopoietic Cell Transplantation: Guidelines from the NMDP/CIBMTR. https://hml.nmdp.org/MacUI/. Accessed December 30, 2019.
  40. 40.↵
    Shaw BE, Logan BR, Spellman SR, et al. Development of an Unrelated Donor Selection Score Predictive of Survival after HCT: Donor Age Matters Most. Biol Blood Marrow Transplant. 2018;24(5):1049–1056. doi:10.1016/j.bbmt.2018.02.006
    OpenUrlCrossRef
  41. 41.↵
    Kollman C, Spellman SR, Zhang MJ, et al. The effect of donor characteristics on survival after unrelated donor transplantation for hematologic malignancy. Blood. 2016;127(2):260–267. doi:10.1182/blood-2015-08-663823
    OpenUrlAbstract/FREE Full Text
  42. 42.↵
    Bosch M, Dhadda M, Hoegh-Petersen M, et al. Immune reconstitution after anti-thymocyte globulin-conditioned hematopoietic cell transplantation. Cytotherapy. 2012;14(10):1258–1275. doi:10.3109/14653249.2012.715243
    OpenUrlCrossRefPubMed
  43. 43.↵
    Farag SS, Fehniger TA, Ruggeri L, Velardi A, Caligiuri MA. Natural killer cell receptors: New biology and insights into the graft-versus-leukemia effect. Blood. 2002;100(6):1935–1947. doi:10.1182/blood-2002-02-0350
    OpenUrlAbstract/FREE Full Text
  44. 44.↵
    Schetelig J, Baldauf H, Heidenreich F, et al. External validation of models for KIR2DS1/KIR3DL1-informed selection of hematopoietic cell donors fails. Blood. 2020;135(16):1386–1395. doi:10.1182/blood.2019002887
    OpenUrlCrossRef
  45. 45.↵
    Cooley S, Trachtenberg E, Bergemann TL, et al. Donors with group B KIR haplotypes improve relapse-free survival after unrelated hematopoietic cell transplantation for acute myelogenous leukemia. Blood. 2009;113(3):726–732. doi:10.1182/blood-2008-07-171926
    OpenUrlAbstract/FREE Full Text
  46. 46.↵
    Weisdorf D, Cooley S, Wang T, et al. KIR B donors improve the outcome for AML patients given reduced intensity conditioning and unrelated donor transplantation. Blood Adv. 2020;4(4):740–754. doi:10.1182/bloodadvances.2019001053
    OpenUrlCrossRef
  47. 47.↵
    Moesta AK, Norman PJ, Yawata M, Yawata N, Gleimer M, Parham P. Synergistic Polymorphism at Two Positions Distal to the Ligand-Binding Site Makes KIR2DL2 a Stronger Receptor for HLA-C Than KIR2DL3. J Immunol. 2008;180(6):3969–3979. doi:10.4049/jimmunol.180.6.3969
    OpenUrlAbstract/FREE Full Text
  48. 48.↵
    Clausen J, Böhm A, Straßl I, et al. HLA-C KIR-ligands determine the impact of anti-thymocyte globulin (ATG) on graft versus host and graft versus leukemia effects following hematopoietic stem cell transplantation. Biomedicines. 2017;5(2). doi:10.3390/biomedicines5020013
    OpenUrlCrossRef
Back to top
Posted July 11, 2020.
Download PDF
Data/Code
Email
Inhibitory KIR-Ligand Interactions and Relapse Protection Following HLA Matched Allogeneic HCT for AML
Elizabeth Krieger, Rehan Qayyum, Amir Toor
medRxiv 2020.07.10.20149047; doi: https://doi.org/10.1101/2020.07.10.20149047
Twitter logo Facebook logo LinkedIn logo Mendeley logo
Citation Tools

Subject Area