Original Article

Bone Marrow Transplantation (2012) 47, 817–823; doi:10.1038/bmt.2011.181; published online 5 December 2011

Post-Transplant Events

Differential impact of inhibitory and activating Killer Ig-Like Receptors (KIR) on high-risk patients with myeloid and lymphoid malignancies undergoing reduced intensity transplantation from haploidentical related donors

D-F Chen1,6, V K Prasad2,6, G Broadwater3, N L Reinsmoen4, A DeOliveira1, A Clark1, K M Sullivan5, J P Chute5, M E Horwitz5, C Gasparetto5, G D Long5, Y Yang5, N J Chao5 and D A Rizzieri5

  1. 1Clinical Transplantation Immunology Laboratory, Department of Pathology, Duke University Medical Center (DUMC), Durham, NC, USA
  2. 2Department of Pediatrics, Blood and Marrow Transplantation Division, DUMC, Durham, NC, USA
  3. 3Department of Biostatistics, DUMC, Durham, NC, USA
  4. 4HLA Laboratory, Cedars-Sinai Health Systems, Los Angeles, CA, USA
  5. 5Division of Cellular Therapy, Department of Medicine, DUMC, Durham, NC, USA

Correspondence: Dr VK Prasad, Department of Pediatric Blood and Marrow Transplantion Division, Duke University Medical Center, Box 3350, 1400 Morreene Road, Durham, NC 27710, USA. E-mail: vinod.prasad@duke.edu

6These authors contributed equally to this work and share first authorship.

Received 24 February 2011; Accepted 8 August 2011
Advance online publication 5 December 2011



The impact of activating KIR (aKIR) and inhibitory KIR (iKIR) on OS, relapse-related mortality (RRM) and acute GVHD (aGVHD) was prospectively studied in 84 adults with high-risk hematologic malignancies receiving reduced intensity conditioning (RIC) T-cell depleted hematopoietic SCT (HSCT) from haploidentical related donors. In this clinical model, freedom from RRM is dependent on GVL effect. Patients were divided into myeloid (n=49) and lymphoid (n=35) malignancy groups. KIR-ligand and ligand-ligand models were studied in both GVH and rejection directions and statistically correlated with outcome measures. In the myeloid group, OS was higher (P=0.009) and RRM was lower (P=0.036) in patients missing HLA-C group2 ligand to donor iKIR. OS was higher if patients had >1 missing ligand (P=0.018). In lymphoid malignancy, missing ligand to donor KIR had no impact on OS or RRM. However, OS was better with donor aKIR 2DS2 (P=0.028). There was a trend towards shorter OS in recipient with KIR 2DS1, 2DS5 and 3DS1, although sample sizes were too small to provide inferential statistics. Findings in lymphoid malignancy patients should be further studied. These results suggest that the absence of appropriate HLA ligands in the recipient to donor iKIR may induce GVL without aGVHD in myeloid malignancy patients undergoing TCD-RIC transplants.


KIR; activating KIR; inhibitory KIR; haploidentical; reduced intensity conditioning (RIC); hematopoietic SCT



The benefit of natural killer (NK) cell mediated GVL effect in haploidentical T-cell–depleted hematopoietic SCT (HSCT) was initially reported after a myeloablative conditioning regimen.1 However, many patients, particularly those of older age and with co-morbidities are unable to tolerate myeloablative conditioning. Use of reduced intensity conditioning (RIC) allows them to access HSCT. Since GVL is largely responsible for anti-leukemic benefits of RIC transplants, this model is well suited for the study of factors like KIR that may influence the GVL effect.5

The KIR gene family currently consists of inhibitory KIRs (iKIRs) (KIR 2DL1, KIR 2DL2, KIR 2DL3, KIR 2DL4, KIR 2DL5A, KIR 2DL5B, KIR 3DL1, KIR 3DL2 and KIR 3DL3) and activating KIRs (aKIRs) (KIR 2DS1, KIR 2DS2, KIR 2DS3, KIR 2DS4, KIR 2DS5 and KIR 3DS1) which are located on chromosome 19 and encode for corresponding receptors on NK cells.6 KIR 2DL1 recognizes HLA-C alleles of the ‘C2 group’ with Lys at position 80 (HLA-CLys80), KIR 2DL2/3 recognizes HLA-C alleles of the ‘C1 group’ with Asn at position 80 (HLA-CAsn80), KIR 3DL1 recognizes ‘Bw4’ alleles, and KIR 3DL2 recognizes HLA-A3 and HLA-A11 alleles.7, 8, 9 It is important to remember that not all KIR genes from an individual's genotype may be expressed nor will all NK cells in that individual carry the corresponding KIR protein on the cell surface. Inhibitory receptors inhibit lysis upon binding the corresponding ligand while stimulation of aKIRs leads to killing by the NK cells.10 In normal individuals, aKIRs and iKIRs mediate killing of virally infected or tumor cells through corresponding signals upon interactions with specific class I HLA molecules that act as their cognate ligands.11

Following HSCT, donor-derived NK cells attack allogeneic cells in the patient if the recipient HLA class I ligands do not sufficiently engage their inhibitory receptors.12 Reduced intensity cytoreduction is expected to spare many host derived immune cells and may allow an anti-graft attack by host NK cells if alloreactivity is present in the reverse direction. HLA ligands for aKIRs are not yet defined but KIR 2DS1 recognizes the C2 group of alleles in EBV–transformed B-lymphocytes13, 14 and leukemia cells.12 The current study examines the differences in OS, relapse-related mortality (RRM) and incidence of Grade II–IV acute GVHD (aGVHD) following TCD haploidentical RIC HSCT in patients with lymphoid and myeloid malignancies with respect to the impact of iKIR and aKIRs in the patients and their donors, ligand to ligand mismatch as well as iKIR to ligand alloreactivity in GVH and rejection directions.



Patient, donor and graft characteristics

Between March 2000 and July 2008, 84 consecutive adult patients with relapsed or refractory hematologic malignancies treated at Duke University Medical Center with T-cell depleted PBSC-transplantation from haploidentical related donors following a RIC regimen were included in this prospective study. Donor and patient DNA samples had to be available to be included in the study. All patients lacked suitable related or unrelated BM donors as well as suitable single cord blood units, or they were not eligible for ablative therapy due to age or co-morbidities. All patients were enrolled in IRB approved protocols and had signed informed consent. Diagnoses are shown in Table 1. The patient and donor pairs were 3/6 (n=41), 4/6 (n=29), 5/6 (n=12) or 6/6 (n=2). HLA-C and -DQB1 revealed further mismatches. The distribution of HLA matching was similar between the myeloid and lymphoid malignancy patients by Yates Chi-square test. Patients were categorized as having lymphoid (n=35) or myeloid (n=49) malignancy (Table 1). The median age was 48.5 years (range, 18–72 years). Myeloid patients were older (51 vs 43 years; P=0.032). In general, the patients were heavily pretreated with multiple regimens. A proportion had undergone prior autologous (n=14; 16.5%) or allogeneic (n=10; 11.7%) transplants. Fifty-one percent (n=43) had relapsed or refractory disease at the time of transplant. There was no statistical difference between the myeloid and the lymphoid groups with respect to the prior transplantation and advanced disease status. These characteristics were similar to those reported in a previous paper from our center that focused on the clinical and immunological recovery in a subgroup of these patients.4 The median duration of follow-up for the whole group and two subgroups are shown in Table 1. The median follow up for surviving patients was 30.4 months (range, 13–77 months). The patient–donor sex matching was similar between the lymphoid and the myeloid groups.

Donors received filgrastim 8μg/kg twice daily starting 5 days before pheresis and continuing until apheresis was completed. All purging was accomplished in vivo after stem cell infusion into the patient. In vitro estimation of the degree of T-cell purging was performed in a subset of patients. A 5mL aliquot of patient serum and pheresed donor cells were mixed into a cell suspension containing 25% serum by volume and incubated for 30minutes at 37°C with 2μg of alemtuzumab followed by a flow analysis for viability of the cellular suspension using the 7-aminoactinomycin-D method.15, 16 Donor lymphocyte infusions were planned for all patients who had the persistence of disease after transplantation, if they did not have severe GVHD. Details of cell collection, in vivo T-cell depletion, estimates of CD34+ and CD3+ cell counts, other graft characteristics, clinical protocol and outcomes on partial cohort is described previously.4, 17


The preparative regimen included i.v. infusions of 5 days of alemtuzumab 20mg/day on days −4 to 0 and 4 days of fludarabine 30mg/m2/day and CY 500mg/m2/day on days −5 to −2. All patients with 3–5/6 HLA-matched grafts received mycophenolate 1gm orally twice daily for 45–60 days after transplantation. Starting on day +1, patients received Filgrastim 5mcg/kg (rounded to nearest vial) until the ANC was >1 × 109/L for 2 days.

HLA and KIR typing

HLA-A, -B, -C, -DRB1 and -DQB1 were performed on each recipient and donor by reversed Sequence-Specific Oligonucleotide Probe (rSSOP, Luminex), Sequence-Specific Primer, or Sequence-Based Typing. High resolution HLA-C typing was performed retrospectively by Sequence-Based Typing when necessary to characterize alleles in the different HLA-C ligand groups. Low resolution KIR typing was performed by using Luminex-rSSOP reagents (One Lambda, Canoga Park, CA, USA) for the presence or absence of 16 KIRs including iKIR genes (KIR 2DL1, KIR 2DL2, KIR 2DL3, KIR 2DL4, KIR 2DL5, KIR 3DL1, KIR 3DL2 and KIR 3DL3), aKIR genes (KIR 2DS1, KIR 2DS2, KIR 2DS3, KIR 2DS4, KIR 2DS5 and KIR 3DS1) and pseudogenes (KIR 2DP1 and KIR 3DP1). KIR genotyping was validated by typing known controls (n=16) from IHIWS panel.

Interpretation of NK alloreactivity

Ligand-ligand (L-L) Model (Ligand Incompatibility Model)

The presence of an epitope of HLA ligand for KIR in the donor and its absence in the patient is assumed to represent potential for donor NK alloreactivity against the patient target cells resulting in alloreactivity in the GVH direction. Conversely, the presence of an epitope of HLA ligand for KIR in the patient and its absence in the donor will result in alloreactivity in the rejection direction. In this study, KIR alloreactivity was measured for HLA-C group one and two, and Bw4 ligand incompatibilities. Each recipient/donor pair was evaluated for L-L incompatibility in both directions. A combined ligand incompatible score was also considered. A positive combined score was assigned when either one or both of HLA-C and -Bw4 ligands were incompatible.

Receptor-ligand Model (Missing Ligand Model)

We measured KIR alloreactivity for missing HLA-C groups one (C1) and two (C2), -Bw4 and/or -A3/11 ligands. Each pair was evaluated for the donor KIR genotype and the lack of corresponding ligands in recipient or vice versa. A combined KIR score was also considered. The combined score was the sum of the number of the KIR genes (2DL1/2/3, 3DL1/2) without ligand.

aKIR receptors

Genotyping was performed for the presence of each aKIR receptor (2DS1, 2DS2, 2DS3, 2DS4, 2DS5 and 3DS1) in donors and patients.

Statistical analysis

Outcome measures used in this study included OS, RRM and aGVHD. aGVHD was scored in all patients by standard criteria18 and only grades II–IV were measured for this analysis. The probabilities of aGVHD were displayed using the cumulative-incidence-function method.19 For aGVHD, patients were evaluable after surviving up to day 14, and death without the event was the competing event while patients were censored on the last day of follow-up. OS was measured from the first day of the transplant to death and censored at the last day of follow-up. RRM was measured from the first day of transplantation to death from relapse or progressive disease, or was censored at the time of death from non relapse cause or date of follow-up for living patients. Kaplan-Meier probability curves20 were used to analyze OS, RRM and acute grades II–IV GVHD by KIR alloreactivity for myeloid malignancy and lymphoid malignancy groups separately, and then with the two groups combined. The differences between groups were compared using the log-rank statistics.21 Central tendencies of age and duration of follow-up were compared between myeloid and lymphoid patients using the Wilcoxon Rank-Sum test. Proportions with pre-transplant therapies and relapse or refractory disease at transplant in the myeloid and lymphoid groups were compared using the χ2-test. All P-values are two-sided. Analyses were completed using the SAS system, version 9.2, and R, version 2.1.1.



L-L incompatibility: impact on OS, RRM and aGVHD

In the GVHD/GVL direction, L-L incompatibility for either HLA-C or HLA-Bw4 was present in 24% (n=12) of myeloid and 48.6% (n=17) of lymphoid malignancy patients. In the rejection direction, incompatibility for either HLA-C or HLA-Bw4 was seen in 24% (n=12) of myeloid and 37.1% (n=13) of lymphoid malignancy patients. Subgroup sample sizes were too small to perform statistical analysis for significant differences in L-L incompatibility on OS, RRM or aGVHD.

Missing recipient ligand for donor's iKIR (receptor-ligand alloreactivity in GVH/GVL direction): impact on OS, RRM and aGVHD

Distributions of incompatibilities for HLA-A3/A11, -Bw4, -C1 and -C2 were similar in the myeloid and lymphoid cohorts. In the myeloid cohort, alloreactivity for (i) HLA- group (C2) was associated with higher OS (P=0.009: Figure 1b) as well as lower RRM (P=0.036: Figure 2b); (ii) Bw4 was associated with higher OS (P=0.047: Figure 1a); and (iii) A3/A11 was associated with lower RRM (P=0.02; Figure 2a). Furthermore, an increasing number of alloreactivities led to higher OS (P=0.018 for two or three; P=0.03 for three categories) as shown in Figures 1c and d. RRM was lower in the presence of incompatibility involving A3/A11 (P=0.02), C2 (P=0.036) or combined (P=0.014) as shown in Figures 2a–c. In contrast, in the lymphoid cohort there was no impact of either individual incompatibility for HLA-A3/A11, -Bw4, -C1 and -C2 or multiple incompatibilities on OS or RRM, but grades II–IV aGVHD trended lower when donor–recipient pairs with C1 or C2 incompatibility (n=24; P=0.011) were analyzed together.

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Impact of receptor-ligand incompatibility in the GVH/GVL direction on OS in patients with myeloid malignancy undergoing T-cell depleted RIC related donor haploidentical transplantation. In this model, incompatibility is defined as the absence of corresponding ligand in the recipient to the iKIR present in the donor. The combinations used were (Figure 1a) KIR 3DL1 for HLA-Bw4 alleles; (Figure 1b) KIR 2DL1 for HLA-C2 alleles that are characterized by lysine at position 80 (HLA-CLys80). Graphs for the impact of receptor-ligand incompatibility representing KIR 2DL2 and KIR 2DL3 for HLA-C1 alleles characterized by Asparagine at position 80 (HLA-CAsn80), and KIR 3DL2 for HLA-A3 and HLA-A11 alleles (not shown because of the lack of statistical significance). Figure 1c depicts the impact of incompatibility for any of the above loci; and Figure 1d shows the impact of multiple incompatibilities.

Full figure and legend (127K)

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Impact of receptor-ligand incompatibility in the GVH/GVL direction on relapse-related mortality (RRM) in patients with myeloid malignancy undergoing T-cell depleted RIC related donor haploidentical transplantation. Similar to Figure 1, incompatibility is defined as the absence of corresponding ligand in the recipient to the iKIR present in the donor. The combinations used were (Figure 2a) KIR 3DL2 for HLA-A3 and HLA-A11 alleles; (Figure 2b) KIR 2DL1 for HLA-C2 alleles that are characterized by lysine at position 80 (HLA-CLys80). Graphs for the impact of receptor-ligand incompatibility representing KIR 3DL1 for HLA-Bw4 alleles, and KIR 2DL2 and KIR 2DL3 for HLA-C1 alleles characterized by asparagine at position 80 (HLA-CAsn80) (not shown because of lack of statistical significance). Figure 2c depicts the combined impact impact of incompatibility for any of the above loci.

Full figure and legend (100K)

Missing donor ligand for recipient's iKIR (receptor–ligand alloreactivity in rejection direction): impact on OS, RRM and aGVHD

Distribution of alloreactivities for HLA-A3/A11, -Bw4, -C1 and -C2 were similar in the myeloid and lymphoid cohorts. In the myeloid cohort, alloreactivity for Bw4 was associated with higher OS (P=0.014). There was no impact of individual or combined alloreactivities on OS or RRM in the lymphoid malignancy cohort.

aKIR in patients and donors

Genotyping was performed for aKIR receptors (2DS1, 2DS2, 2DS3, 2DS4, 2DS5 and 3DS1) in donors and patients. However, for the outcome analysis, KIR 2DS4 was excluded because almost all samples were positive for it. The presence or absence of any of the aKIRs in either recipient or donor had no impact on OS, RRM or GVHD in myeloid patients. However, in lymphoid patients, a trend towards higher GVHD probability was observed if donor was positive for aKIR 2DS1 (n=9; P=0.031). OS in lymphoid patients tended to be higher if the donor carried aKIR 2DS2 (P=0.028; n=16) and lower if the recipient carried aKIR 2DS1 (n=10; P=0.011), KIR 2DS5 (n=9; P=0.035) or KIR 3DS1 (n=10; P=0.011).



Despite numerous studies in the last decade investigating the impact of KIR on the outcomes of HSCT, a cohesive, all-encompassing and clear theory with supportive clinical evidence regarding the relevant associations has yet to emerge. These associations have been studied separately in various donor sources including matched sibling and unrelated donors, graft sources including BM, PBSCs and umbilical cord blood, and following various cytoreductive regimens with or without T-cell depletion.22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32 More recently, cellular and molecular studies have elucidated the complexity of the KIR genes, its ligand interactions and effector responses, and allowed the more recent outcome analyses to incorporate a wider range of variables and models of the donor–host NK cell alloreactivities.10, 27, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42 Our study includes patients transplanted at a single center using uniform conditioning, graft source, GVHD prophylaxis, supportive care and follow-up on a prospective basis. In this cohort, a high degree of T-cell depletion created a non-competing environment with a low GVHD and minimal need for additional immunosuppression and thus allowed for a clearer evaluation of donor and recipient NK interactions and effector responses. Post transplantation lymphocytic recovery patterns highlighting relative maintenance of the CD56+ compartment (including NK cells) relative to the CD4 and CD8 compartments with this type of conditioning regimen have been documented in our previous work.4 To our knowledge, this is the first single center analysis of iKIR and aKIR genotype and ligand in RIC transplants, a group that is arguably more dependent on GVL for OS and freedom from relapse related mortality.

Receptor-Ligand alloreactivity in the GVL/GVH direction in our cohort revealed the greatest degree of impact on the outcome measures. Absence of recipient ligand for donor iKIR (GVH/GVL direction) impacted OS and RRM in the myeloid patients (Figures 1 and 2). Alloreactivity involving HLA-C2 was most significant and led to higher OS as well as lower RRM. A functional study by Pende et al.12 provides evidence that KIR 2DL1 can discriminate between C1 and C2 groups leading to differential lysis of C1 vs C2 allele-expressing leukemia blasts. The additional impact of receptor-ligand alloreactivity in our myeloid patients was reflected in a higher OS in relation to Bw4 and lower RRM in relation to A3/A11. The presence of multiple alloreactivities led to the highest OS and lowest RRM. We also noticed a difference between HLA-C1 and HLA-C2 alleles. Myeloid patients with alloreactivity involving HLA-C1 did not have an advantage in OS and RRM. In fact, C1 alloreactivity resulted in higher grades II–IV aGVHD. Cook et al.43 showed similar findings in matched sibling transplants if the donor carried aKIR 2DS2. KIR ‘L-L’ incompatibility did not predict OS, RRM or aGVHD in our study. It has been shown that persistence of T-cells inhibits NK cell alloreactivity.35 It is unclear if that was a factor in this cohort. Similar to our observations, Gagne et al.27 did not find L-L analysis to be very useful in predicting survival and relapse following unrelated donor transplants. Overall, our findings support the view that receptor-ligand model alloreactivity (with the exception of HLA-C alleles of group C1) in the GVH/GVL direction in myeloid patients induces a GVL effect without causing GVHD and thus leads to a lower RRM and higher OS.

The lower probability of grades II–IV aGVHD in lymphoid malignancy (n=24) in the presence of mismatch in the receptor-ligand model is previously unreported and merits further scrutiny. Higher GVHD was reported in the presence of receptor-ligand mismatch in two studies done in myeloablative transplant recipients; one from related haploidentical donors44 and another using T-cell replete grafts from unrelated donors.23 In a RIC cord blood transplant study, KIR-ligand alloreactivity resulted in significantly higher rates of grade III–IV aGVHD (42% vs 13%) and TRM (27% vs 12%) with inferior survival (32% vs 52%).24 A recent study provides compelling evidence that the expression of mature KIR repertoires is affected by T-cells in the graft which in turn impacts clinical outcomes after unrelated transplantation.35 In that context, it is noteworthy that we observed some impact of aKIR in this study. It is important to keep in mind that the conditioning regimen, GVHD prophylaxis, T-cell dose and infused NK cells, graft sizes and T-cell depletion protocols will differ between studies and institutions and may be impacting NK mediated effects. Additionally, in lymphoid patients, a higher GVHD probability was seen if the donor was positive for aKIR 2DS1. OS in lymphoid patients was higher if the donor carried aKIR 2DS2 and lower if the recipient carried aKIR 2DS1, KIR 2DS5 or KIR 2DS1. These findings must be interpreted cautiously because of small numbers and further explored in larger patient populations. Interestingly, Pende et al.12 showed that leukemic blasts expressing C2/C2 ligands were killed by KIR 2DL2/3+ NK cells that co-expressed KIR 2DS1 because aKIR 2DS1 could directly recognize C2 on leukemia cells. In the large study of 1087 patients transplanted using unrelated donors through NMDP, recipients of KIR 3DS1 positive donors had lower grade II–IV acute GVHD but no impact on relapse.29

In conclusion, absence of appropriate HLA ligands in the recipient to donor iKIR may induce GVL without aGVHD in myeloid malignancy patients undergoing haploidentical T-cell depleted RIC transplant from haploidentical related donor. Similar alloreactivity in lymphoid malignancy patients did not have a GVL effect although impact on aGVHD was noted. Cumulative evidence from all previous studies and the current study points to the level of uniqueness of NK mediated effects in specific contexts and suggests that clinical applicability of NK incompatibility must be determined and utilized separately for specific donor sources, conditioning regimens, levels of HLA matching, graft manipulation and disease characteristics instead of extrapolating across these variables.


Conflict of interest

The authors declare no conflict of interest.



  1. Ruggeri L, Capanni M, Urbani E, Perruccio K, Shlomchik WD, Tosti A et al. Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic transplants. Science 2002; 295: 2097–2100. | Article | PubMed | ISI | CAS |
  2. Hale G, Zhang MJ, Bunjes D, Prentice HG, Spence D, Horowitz MM et al. Improving the outcome of bone marrow transplantation by using CD52 monoclonal antibodies to prevent graft-versus-host disease and graft rejection. Blood 1998; 92: 4581–4590. | PubMed | ISI | CAS |
  3. Morris E, Thomson K, Craddock C, Mahendra P, Milligan D, Cook G et al. Outcomes after alemtuzumab-containing reduced-intensity allogeneic transplantation regimen for relapsed and refractory non-Hodgkin lymphoma. Blood 2004; 104: 3865–3871. | Article | PubMed | ISI | CAS |
  4. Rizzieri DA, Koh LP, Long GD, Gasparetto C, Sullivan KM, Horwitz M et al. Partially matched, nonmyeloablative allogeneic transplantation: clinical outcomes and immune reconstitution. J Clin Oncol 2007; 25: 690–697. | Article | PubMed | ISI | CAS |
  5. Barrett AJ. Understanding and harnessing the graft-versus-leukaemia effect. Br J Haematol 2008; 142: 877–888. | Article | PubMed | ISI | CAS |
  6. Trowsdale J. Genetic and functional relationships between MHC and NK receptor genes. Immunity 2001; 15: 363–374. | Article | PubMed | ISI | CAS |
  7. Moretta A, Bottino C, Vitale M, Pende D, Biassoni R, Mingari MC et al. Receptors for HLA class-I molecules in human natural killer cells. Annu Rev Immunol 1996; 14: 619–648. | Article | PubMed | ISI | CAS |
  8. Lanier LL. NK cell receptors. Annu Rev Immunol 1998; 16: 359–393. | Article | PubMed | ISI | CAS |
  9. Long EO. Regulation of immune responses through inhibitory receptors. Annu Rev Immunol 1999; 17: 875–904. | Article | PubMed | ISI | CAS |
  10. Long EO. Immunology Signal sequences stop killer cells. Nature 1998; 391: 740–743. | Article | PubMed | ISI |
  11. Vivier E, Tomasello E, Baratin M, Walzer T, Ugolini S. Functions of natural killer cells. Nat Immunol 2008; 9: 503–510. | Article | PubMed | ISI | CAS |
  12. Pende D, Marcenaro S, Falco M, Martini S, Bernardo ME, Montagna D et al. Anti-leukemia activity of alloreactive NK cells in KIR ligand-mismatched haploidentical HSCT for pediatric patients: evaluation of the functional role of activating KIR and redefinition of inhibitory KIR specificity. Blood 2009; 113: 3119–3129. | Article | PubMed | ISI | CAS |
  13. Moretta A, Sivori S, Vitale M, Pende D, Morelli L, Augugliaro R et al. Existence of both inhibitory (p58) and activatory (p50) receptors for HLA-C molecules in human natural killer cells. J Exp Med 1995; 182: 875–884. | Article | PubMed | ISI | CAS |
  14. Stewart CA, Laugier-Anfossi F, Vely F, Saulquin X, Riedmuller J, Tisserant A et al. Recognition of peptide-MHC class I complexes by activating killer immunoglobulin-like receptors. Proc Natl Acad Sci U S A 2005; 102: 13224–13229. | Article | PubMed | ISI |
  15. Gratama JW, Sutherland DR, Keeney M. Flow cytometric enumeration and immunophenotyping of hematopoietic stem and progenitor cells. Semin Hematol 2001; 38: 139–147. | Article | PubMed | ISI |
  16. Sutherland DR, Anderson L, Keeney M, Nayar R, Chin-Yee I. The ISHAGE guidelines for CD34+ cell determination by flow cytometry. International Society of Hematotherapy and Graft Engineering. J Hematother 1996; 5: 213–226. | Article | PubMed | CAS |
  17. Rizzieri DA, Dev P, Long GD, Gasparetto C, Sullivan KM, Horwitz M et al. Response and toxicity of donor lymphocyte infusions following T-cell depleted non-myeloablative allogeneic hematopoietic SCT from 3-6/6 HLA matched donors. Bone Marrow Transplant 2009; 43: 327–333. | Article | PubMed | ISI |
  18. Przepiorka D, Weisdorf D, Martin P, Klingemann HG, Beatty P, Hows J et al. 1994 consensus conference on acute GVHD grading. Bone Marrow Transplant 1995; 15: 825–828. | PubMed | ISI | CAS |
  19. Gooley TA, Leisenring W, Crowley J, Storer BE. Estimation of failure probabilities in the presence of competing risks: new representations of old estimators. Stat Med 1999; 18: 695–706. | Article | PubMed | ISI | CAS |
  20. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. JAm Stat Assoc 1958; 53: 457–481. | Article |
  21. Mantel C, Luo Z, Hendrie P, Broxmeyer HE. Steel factor and granulocyte-macrophage colony stimulating factor act together to enhance choline-lipid turnover during synergistically stimulated proliferation of the human factor dependent cell line, M07E. Biochem Biophys Res Commun 1993; 97: 978–984. | Article |
  22. Giebel S, Locatelli F, Lamparelli T, Velardi A, Davies S, Frumento G et al. Survival advantage with KIR ligand incompatibility in hematopoietic stem cell transplantation from unrelated donors. Blood 2003; 102: 814–819. | Article | PubMed | ISI | CAS |
  23. Davies SM, Ruggieri L, DeFor T, Wagner JE, Weisdorf DJ, Miller JS et al. Evaluation of KIR ligand incompatibility in mismatched unrelated donor hematopoietic transplants. Blood 2002; 100: 3825–3827. | Article | PubMed | ISI | CAS |
  24. Brunstein CG, Wagner JE, Weisdorf DJ, Cooley S, Noreen H, Barker JN et al. Negative effect of KIR alloreactivity in recipients of umbilical cord blood transplant depends on transplantation conditioning intensity. Blood 2009; 113: 5628–5634. | Article | PubMed | ISI |
  25. Cooley S, Trachtenberg E, Bergemann TL, Saeteurn K, Klein J, Le CT et al. Donors with group B KIR haplotypes improve relapse-free survival after unrelated hematopoietic cell transplantation for acute myelogenous leukemia. Blood 2009; 113: 726–732. | Article | PubMed | ISI | CAS |
  26. Hsu KC, Keever-Taylor CA, Wilton A, Pinto C, Heller G, Arkun K et al. Improved outcome in HLA-identical sibling hematopoietic stem-cell transplantation for acute myelogenous leukemia predicted by KIR and HLA genotypes. Blood 2005; 105: 4878–4884. | Article | PubMed | ISI | CAS |
  27. Gagne K, Busson M, Bignon JD, Balere-Appert ML, Loiseau P, Dormoy A et al. Donor KIR3DL1/3DS1 gene and recipient Bw4 KIR ligand as prognostic markers for outcome in unrelated hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 2009; 15: 1366–1375. | Article | PubMed | ISI |
  28. Farag SS, Bacigalupo A, Eapen M, Hurley C, Dupont B, Caligiuri MA et al. The effect of KIR ligand incompatibility on the outcome of unrelated donor transplantation: a report from the center for international blood and marrow transplant research, the European blood and marrow transplant registry, and the Dutch registry. Biol Blood Marrow Transplant 2006; 12: 876–884. | Article | PubMed | ISI | CAS |
  29. Venstrom JM, Gooley TA, Spellman S, Pring J, Malkki M, Dupont B et al. Donor activating KIR3DS1 is associated with decreased acute GVHD in unrelated allogeneic hematopoietic stem cell transplantation. Blood 2010; 115: 3162–3165. | Article | PubMed | ISI |
  30. Giebel S, Nowak I, Dziaczkowska J, Czerw T, Wojnar J, Krawczyk-Kulis M 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: 343–356. | Article | PubMed | ISI | CAS |
  31. Locatelli F, Pende D, Maccario R, Mingari MC, Moretta A, Moretta L. Haploidentical hemopoietic stem cell transplantation for the treatment of high-risk leukemias: how NK cells make the difference. Clin Immunol 2009; 133: 171–178. | Article | PubMed | ISI | CAS |
  32. Willemze R, Rodrigues CA, Labopin M, Sanz G, Michel G, Socie G et al. KIR-ligand incompatibility in the graft-versus-host direction improves outcomes after umbilical cord blood transplantation for acute leukemia. Leukemia 2009; 23: 492–500 [Erratum appears in Leukemia. 2009 Mar;23(3):630]. | Article | PubMed | ISI | CAS |
  33. Caligiuri MA. Human natural killer cells. Blood 2008; 112: 461–469. | Article | PubMed | ISI | CAS |
  34. Cambiaggi A, Darche S, Guia S, Kourilsky P, Abastado JP, Vivier E. Modulation of T-cell functions in KIR2DL3 (CD158b) transgenic mice. Blood 1999; 94: 2396–2402. | PubMed | ISI | CAS |
  35. Cooley S, McCullar V, Wangen R, Bergemann TL, Spellman S, Weisdorf DJ et al. KIR reconstitution is altered by T cells in the graft and correlates with clinical outcomes after unrelated donor transplantation. Blood 2005; 106: 4370–4376. | Article | PubMed | ISI | CAS |
  36. Fehniger TA, Carson WE, Caligiuri MA. Costimulation of human natural killer cells is required for interferon gamma production. Transplant Proc 1999; 31: 1476–1478. | Article | PubMed | ISI |
  37. Middleton D, Gonzelez F. The extensive polymorphism of KIR genes. Immunology 2010; 129: 8–19. | Article | PubMed | ISI |
  38. Miller JS. How killers kill. Blood 2008; 112: 213. | Article | PubMed | ISI |
  39. Ruggeri L, Mancusi A, Capanni M, Urbani E, Carotti A, Aloisi T et al. Donor natural killer cell allorecognition of missing self in haploidentical hematopoietic transplantation for acute myeloid leukemia: challenging its predictive value. Blood 2007; 110: 433–440. | Article | PubMed | ISI | CAS |
  40. Shilling HG, McQueen KL, Cheng NW, Shizuru JA, Negrin RS, Parham P. Reconstitution of NK cell receptor repertoire following HLA-matched hematopoietic cell transplantation. Blood 2003; 101: 3730–3740. | Article | PubMed | ISI | CAS |
  41. Velardi A, Ruggeri L, Mancusi A, Aversa F, Christiansen FT. Natural killer cell allorecognition of missing self in allogeneic hematopoietic transplantation: a tool for immunotherapy of leukemia. Curr Opin Immunol 2009; 21: 525–530. | Article | PubMed | ISI |
  42. Yu J, Venstrom JM, Liu XR, O’Reilly R, Pring J, Hasan RS et al. Breaking tolerance to self, circulating natural killer cells expressing inhibitory KIR for non-self HLA exhibit effector function following T-cell depleted allogeneic hematopoietic cell transplantation. Blood 2009; 113: 3875–3884. | Article | PubMed | ISI | CAS |
  43. Cook MA, Milligan DW, Fegan CD, Darbyshire PJ, Mahendra P, Craddock CF et al. The impact of donor KIR and patient HLA-C genotypes on outcome following HLA-identical sibling hematopoietic stem cell transplantation for myeloid leukemia. Blood 2004; 103: 1521–1526. | Article | PubMed | ISI | CAS |
  44. Bishara A, De SD, Witt CC, Brautbar C, Christiansen FT, Or R et al. The beneficial role of inhibitory KIR genes of HLA class I NK epitopes in haploidentically mismatched stem cell allografts may be masked by residual donor-alloreactive T cells causing GVHD. Tissue Antigens 2004; 63: 204–211. | Article | PubMed | ISI | CAS |


NJC is supported by PO1 (NJC) 2PO1-CA047741-16A1 from NIH; DAR is a Leukemia Lymphoma Society Scholar in Clinical Research.

Author contribution: VKP conceptualized and designed the study, collected clinical data, analyzed the data, wrote the manuscript and led all the coordination efforts; DAR and NJC were involved with conceptualizing, designing and overseeing the study, development of the transplant protocol, collection of clinical data and manuscript preparation; DFC was involved in conceptualizing the study, generation of HLA and KIR typing data in the laboratory, data analysis and editing the manuscript; GB was involved in statistical analysis, review of the data and manuscript preparation; NLR was involved with conceptualizing the study, laboratory support and manuscript preparation; ADO and AC contributed to laboratory studies; KMS, JPC, MEH, CG and GDL were involved with the collection of clinical data and manuscript preparation; YY contributed in manuscript preparation.