- Email: [email protected]
Is there an impact of killer cell immunoglobulin-like receptors and KIR-ligand incompatibilities on outcomes after unrelated cord blood stem cell transplantation?
Best Practice & Research Clinical Haematology 23 (2010) 283–290
Contents lists available at ScienceDirect
Best Practice & Research Clinical Haematology journal homepage: www.elsevier.com/locate/beha
13
Is there an impact of killer cell immunoglobulin-like receptors and KIR-ligand incompatibilities on outcomes after unrelated cord blood stem cell transplantation? Roel Willemze, MD *, Annalisa Ruggeri, MD, Duncan Purtill, MD, Celso Arrais Rodrigues, MD, Eliane Gluckman, MD, Vanderson Rocha on behalf of Eurocord and of the European Group of Blood and Marrow Transplantation Department of Hematology, Leiden University Medical Center, Albinusdreef 2, Leiden, the Netherlands
Keywords: KIR KIR-ligand incompatibility umbilical cord blood transplantation
Donor killer cell immunoglobulin-like receptor (KIR) ligand incompatibility in the graft-versus-host direction is associated with decreased relapse incidence and improved disease-free survival after haploidentical and human leucocyte antigen (HLA)mismatched unrelated, haematopoietic stem cell transplantation. However, review of all published studies of allogeneic HLAmatched or mismatched stem cell transplantation shows that the results on the relationship between donor–recipient KIR(-ligand) (in)compatibility and outcomes are highly variable, ranging from highly beneficial to detrimental. Reasons for these differences may include the methodology to determine KIR(-ligand) incompatibility, the disease distribution and the transplant protocol or donor type. Two retrospective studies on the effects of KIR-ligand incompatibility in unrelated cord blood transplantation (UCBT) for haematological malignancies have resulted in conflicting results. The Eurocord study showed a favourable effect of KIR-ligand mismatching on relapse incidence and leukaemia-free survival, whereas the Minneapolis study showed no effect on these end points and a detrimental effect on incidence of graft-versus-host disease (GvHD). In patients with non-malignant disorders, KIRligand (in)compatibility between donor and recipient was not associated with outcomes in a recent Eurocord analysis. Therefore,
* Corresponding author. E-mail address: [email protected] (R. Willemze). 1521-6926/$ – see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.beha.2010.05.005
284
R. Willemze et al. / Best Practice & Research Clinical Haematology 23 (2010) 283–290
the role of natural killer (NK) cell alloreactivity in UCBT is far from clear. It is too early to use a donor–recipient KIR(-ligand) algorithm for selection of a cord blood donor. Ó 2010 Elsevier Ltd. All rights reserved.
Allogeneic haematopoietic stem cell transplantation Allogeneic human leucocyte antigen-(HLA) matched or mismatched haemato- poietic stem cell transplantation (allo-SCT) has been applied for decades as a method of stem cell reconstitution in patients who have received myeloablative chemotherapy and irradiation for the treatment of haematological diseases. Usually, the origin of the stem cell graft is either bone marrow or mobilised stem cells from peripheral blood. Since 15 years, umbilical cord blood units from unrelated donors constitute another well-established source of stem cells for haematopoietic stem cell transplantation. Initially, the only purpose of stem cell transplantation was to replace the irreversibly damaged haematopoietic stem cell compartment after chemotherapy and radiotherapy at the maximum doses that could be tolerated by non-haematopoietic tissues. However, in contrast to autologous stem cell transplantation, allo-SCT is accompanied by extra immunological benefits, such as antitumour response, and complications such as graft-versus-host disease (GvHD), immune incompetence or graft rejection. It became apparent that donor-derived T-cells play an important role in several of these complications, as well as in the incidence of recurrence of the original haematological disease by recognising differences in minor and major histocompatibility antigens between the donor and the recipient [1,2]. More recently, a novel immunological mechanism with clinical implications after allogeneic stem cell transplantation that relies on interactions between donor natural killer (NK) cells and recipient cells, was demonstrated in the setting of haploidentical family donor SCT [3–5].
Natural Killer (NK) cells NK cells are large granular lymphocytes originating in the bone marrow from CD34þ cells. Owing to their early production of cytokines and chemokines (interferon(IFN)-g, tumour necrosis factor (TNF)a and granulocyte macrophage-colony stimulating factor (GM-CSF) and their ability to lyse target cells without prior sensitisation, NK cells (10–15% of all lymphocytes; CD56þ, CD3) are, as innate immune lymphocytes, critical to host defence against invading infectious pathogens and malignant transformation [6]. NK cells can be distinguished from other lymphocytes by the expression of the surface markers CD56 and CD16 and the absence of CD3. Based on the cell surface density of CD56 and CD16, human NK cells can be divided into two subsets, the regulatory and the cytotoxic NK cells [7]. The immunoregulatory CD56bright, CD16dim or negative cells (10% of NK cells) are large agranular cells, that express intermediate and high-affinity immunoglobulin-like (IL)-2, and high CD94/NKG2 and low killer cell immunoglobulin-like receptor (KIR) receptors, and that secrete, upon exogenous stimulation with pro-inflammatory cytokines, a greater amount of cytokines and chemokines than CD56dim NK cells [8,9]. The cytotoxic CD56dim, CD16 positive cells (90% of NK cells) are cells with the cytoplasm containing cytolytic granules that express intermediate affinity for IL-2, high CD94/NKG2 and high KIR receptors and are the effectors of natural cell lysis or antibody-dependent cellular cytotoxicity (ADCC) [8,9]. Recently, it was shown that in response to target recognition, these cells, rather than CD56bright NK cells, are primarily responsible for production of cytokines and chemokines [10]. The mechanism behind NK cell cytolytic activity was first proposed by Ljunggren and Karre[11] in 1990 as the ‘missing self’ hypothesis. They suggested that NK cells kill target cells when they lack or have a low expression of ‘self’ HLA class I molecules. The molecular proof came in 1992 with the identification of inhibitory NK-cell receptors [12]. Since then, a wide range of both inhibitory and activating receptors that regulate NK-cell function has been identified [13–15].
R. Willemze et al. / Best Practice & Research Clinical Haematology 23 (2010) 283–290
285
Killer cell immunoglobulin-like receptors The cytolytic activity of human NK cells is modulated by the interaction of two types of structurally different, inhibitory and activating, major histocompatibility complex (MHC) Class 1 specific receptor superfamilies, the CD94–NKG2 family and the killer cell immunoglobulin-like receptors (KIR) family, with MHC class 1 antigens expressed by host cells. The CD94–NKG2 family comprises the C-type lectinslike receptors, which form heterodimers of the invariant CD94 chain and different members of the NKG2 family. These receptors are encoded on chromosome 12. The inhibitory CD94–NKG2A receptors fulfil an important function in the prevention of NK cell autoreactivity. They recognise the non-classical HLA-E class Ib molecules. The immunoglobulin-like receptor family includes the KIRs. KIRs are encoded on chromosome 19 and are expressed on NK cells and a subset of T-cells. They play an important role in the prevention of autoreactivity and the elimination of viral-infected and tumour cells [13–15]. The KIRs are transmembrane receptors encoded by 17 genes. The nomenclature of these genes is based on three structural criteria: the number of extracellular immunoglobulin domains that varies between two (2D) or three (3D) domains, the presence of a short (S) or long (L) intracellular domain (cytoplasmic tail) and the sequence similarity [16]. These criteria define the functionally relevant KIR2DL1-5, KIR3DL1-3, KIR2DS1-5 and KIR3DS1. The structural characteristics correlate with function. The length of the cytoplasmic tail conveys the type of signal that is generated upon ligand binding. The KIRs with a long cytoplasmic tail inhibit NK cell reactivity by recruitment of the phosphatase Src homology 2 domain-containing phosphatase 1 (SHP-1) to the receptors’ phosphorylated cytoplasmic immunoreceptor thyrosine-based inhibition motifs (ITIMs). KIRs with a short cytoplasmic tail lack these ITIMs but have a positively charged lysine residue in the transmembrane domain, which associates with the DAP12 signalling molecule and thereby induces NK reactivity [15]. KIR ligands The HLA class I molecules act as ligands for some of the KIR genes. The majority of KIRs is specific for HLA-C, but HLA-A, -B and -G have also been described as ligands for some KIRs. KIRs do not recognise each of the allelic variants of the HLA-C molecule, but are restricted to two known epitope groups of HLA-C. HLA-C group 1 (C1) covers the alleles with an asparagine at position 80, while the group 2 (C2) alleles have a lysine at that position. Each of both epitope groups are recognised by inhibitory and activating KIRs [17]. The inhibitory KIR2DL2/KIR2DL3 receptors and the activating KIR2DS2 recognise HLA-C group 1 epitopes, and the inhibitory KIR2DL1 and the activating KIR2DS1 receptor recognise HLA-C group 2 epitopes. KIR3DL1 and KIR3DS1 are inhibitory and activating receptors, respectively, for HLA-B allotypes with Bw4 motifs at positions 77–83. Recently, the specificity of KIR3DL1 has been found to be restricted to all Bw4 alleles (except for 1301 and 1302) and to some HLA-A alleles such as 2301, 2402 and 3201[18,19]; KIR2DL4 recognises HLA-G. KIR3DL2 has as its ligand HLA-A3 and HLA-A11 allele families, but only when certain virally derived peptides are loaded [20]. Donor-versus-recipient NK cell alloreactivity in allo-SCT Donor NK cells generate alloreactions against recipient cells through their KIR receptors when they are mismatched with the recipient for the HLA-C groups and/or the HLA-Bw4 group. Furthermore, HLA-A3 and HLA-A11 function as ligands for KIR3DL2 presumably only when binding EBV reactivation in vitro.Therefore, there is some controversy on the use of HLA-A3 and HLA-A11 as KIR ligand in the clinical setting [20]. The genes encoding inhibitory KIR are nearly always present in populations at frequencies greater than 90%. These genes are not necessarily expressed as corresponding receptor by the NK cell and if expressed, surface expression levels and strength of inhibitory signal may vary considerably [21–23]. Functional analysis of NK clones against HLA-C group 1 or group 2, or HLA-Bw4 tested with cells from mismatched recipients showed alloreactivity in virtual all HLA-C mismatched cases tested, but only in a number of HLA-Bw4 mismatched combinations. No donor alloreactive NK clones could be detected in donor–recipient combinations without KIR-ligand mismatches in the GvH direction, indicating that KIR-ligand mismatching may be a prerequisite for NK cell alloreactivity [24].
286
R. Willemze et al. / Best Practice & Research Clinical Haematology 23 (2010) 283–290
Donor-versus-recipient NK cell alloreactivity and allogeneic blood or bone-marrow-derived stem cell transplantation The role of KIRs and KIR-ligand in the context of allogeneic blood or bone marrow stem cell transplantation has been investigated over the past 10 years. In transplants that are KIR-ligand mismatched in the GvH direction, donor NK cells expressing inhibitory KIRs, which do not recognise ligand(s) on recipient targets, are released from HLA inhibition and mediate allo-responses leading to clinically significant effects. Clinical proof for these NK cell-induced effects were reported by Ruggeri et al. [25,26], who, using haplo-identical family donor SCT, has shown that KIR or KIR-ligand incompatibility in the GvH direction was associated with a decreased incidence of graft failure, acute GvHD and relapse and an improved survival of patients with acute myeloid leukaemia (AML). They explain their clinical results by findings from mouse and human experiments in which donor-versus-recipient alloreactive NK cells eliminate residual leukaemic cells, ablate host-type dendritic cells responsible for triggering GvHD, and attack residual host lympho–haematopoietic cells including T-cells that are responsible for graft rejection [26]. Different algorithms have been used to investigate donor-versus-recipient NK-cell alloreactivity after allo-SCT. The most reliable method to determine NK-cell alloreactivity is probably by donor KIR clones against KIR-ligands on recipient cells or by donor KIR genotyping. However, since data on KIR genotype are usually not readily available, the most frequently applied algorithm is the usage of the presence of KIR-ligand compatibility or incompatibility between the donor and the recipient. These algorithms have applied to investigate the clinical relevance of donor–recipient NK-cell alloreactivity in a number of series of HLA-matched or mismatched related- or unrelated stem cell transplantations. The results of these studies with respect to the outcome of allogeneic SCT were highly variable, ranging from highly beneficial to detrimental [27–37]. These differences in outcomes may be partly due to many differences between the studies. First, as already mentioned, the methods to determine donor-versus-recipient NK-cell alloreactivity varied. Furthermore, some authors included only HLA-C and HLA-B, others also included HLA-A KIR-ligands; and in some studies inhibitory as well as activating KIR receptors were included in the analysis [36,37]. Second, inclusion of patients of all ages with various haematological malignancies in different stages of the disease, transplanted with stem cells from HLA-matched as well as mismatched, family or unrelated donors after various kinds of conditioning regimens may have influenced outcomes with respect to NK-cell reactivity. Third, the number of donor or recipient T-cells involved in the transplantation procedure differed among the studies. T- and NK-cells of the donor have been shown to play a role in HLA-matched as well as mismatched haematopoietic stem cell transplantation (HSCT). To what extent both alloreactivities influence each other in the clinical setting, is not extensively studied. Lowe et al. [38] suggested that Tcell alloreactivity may mask NK-cell alloreactivity in minimally T-cell-depleted grafts of HLA-nonidentical donors. The published studies fail to show a clear pattern with respect to T-cell depletion and the effects of NK-cell alloreactivity. Extensive T-cell depletion by CD34þcell selection and intravenous anti-thymocyte or anti-lymphocyte globulin (ATG/ALG) in the conditioning regimen has only been performed in the haploidentical setting [24–26], a schedule that is associated with rapid recovery of NK cells and slow T-cell recovery, but with the strongest effects of donor-versus-recipient NK reactivity. In the remaining studies where ATG in the conditioning regimen may have been responsible for some in vivo T-cell depletion, conflicting results are reported. Donor-versus-recipient NK cell alloreactivity and unrelated cord blood stem cell transplantation for patients with malignant and non-malignant diseases Umbilical cord blood stem cell transplantation represents an increasingly popular method of allogeneic HSCT. The main advantages are rapid availability and the possibility to perform HLA-mismatched transplantation, due to a relatively low risk of severe GvHD. This increases considerably the probability of finding a suitable donor, since one or two HLA disparities (on the basis of antigen level HLA-A and HLA-B and allele level HLA-DRB1 typing) between donor and recipient lead to similar outcomes and only more than two HLA disparities are associated with diminished neutrophil recovery, increased non-relapse mortality and decreased survival chances [39,40].
R. Willemze et al. / Best Practice & Research Clinical Haematology 23 (2010) 283–290
287
Moreover, outcomes after unrelated cord blood transplantation (UCBT) are comparable with those after unrelated bone marrow transplantation in children and adults with acute leukaemia [41–43]. Only two retrospective studies have been currently published investigating the possible effects of donor-versus-recipient NK cell alloreactivity after UCBT [44,45]. The Eurocord–Netcord and Acute Leukemia Working Party of the EBMT [45] assessed, using the Eurocord Registry, NK cell alloreactivity by means of KIR-ligand (in)compatibility in the GvH direction in 218 patients with acute leukaemia (AML n ¼ 94, acute lymphoblastic leukaemia (ALL) n ¼ 124) in complete remission, who had received a single unit UCBT from a KIR-ligand compatible or KIR-ligandincompatible donor. Patients and donors were categorised to their degree of KIR-ligand compatibility by determining whether or not they expressed HLA-C group 1 or 2, HLA-Bw4 or HLA-A3/-A11. Sixtynine donor–patient combinations were KIR-ligand mismatched in the GvH direction and 149 were matched. They found that, with a median follow-up time of only 14 months, KIR-ligand incompatibility in the GvH direction between donor and recipient was associated with a significantly reduced incidence of relapse (2 years cumulative incidence 20% vs. 37%, p ¼ 0.03), improved leukaemia-free survival (55% vs. 31%, p ¼ 0.005) and overall survival (OS) (57% vs. 40%, p ¼ 0.02). No effect was found on engraftment pattern, acute or chronic GvHD and non-relapse mortality. These results on relapse incidence, leukaemia-free survival and OS were even more evident for AML transplant recipients [45]. A recently performed update of the data of this series of patients by Annalisa Ruggeri did not show, after a median follow up of 34 months, any significant change in the outcomes. By contrast, Brunstein et al. [44] reported a lack of, or even detrimental, effects of KIR-ligand alloreactivity on acute graft-versus-host disease (aGvHD), relapse incidence (RI) and OS, in 257 recipients of a single (n ¼ 91) or a double (n ¼ 166) unit UCBT following myeloablative (n ¼ 155) or reduced intensity (n ¼ 102) conditioning regimens. After myeloablative conditioning, KIR-ligand incompatibility had no effect on grade II–IV acute GvHD, transplant-related mortality, relapse and survival. By contrast, following reduced intensity conditioning, KIR-ligand incompatibility resulted in higher rates of acute GvHD (42% vs. 13%, p < 0.01) and treatment related mortality (27% vs. 12%, p ¼ 0.03) with inferior survival (32% vs. 52%, p ¼ 0.03). There were no apparent methodological differences between the two analyses with respect to the algorithm of KIR-ligand incompatibility, since the Minnesota group included, just as the Eurocord group, HLA-C groups 1 and 2, HLA-Bw4 as well as the controversial HLA-A3/11 as KIR-ligand for their calculations. In addition, the frequency of KIR-ligand incompatibility in the GvH direction was equal in the two studies (29% vs. 32%, respectively). There are a number of differences with respect to median age, diagnoses, HLA-matching, conditioning regimens, use of ATG and the number of umbilical cord blood (UCB) units used. Diagnosis as explanation can be ruled out since comparing the outcomes for AML only as reported by the Minnesota (n ¼ 88; 62 KIR-ligand matched, 26 mismatched) and by the Eurocord group (n ¼ 94; 68 KIR-ligand matched, 26 mismatched) showed that the Minneapolis group did not see significant effects of KIR-ligand incompatibility, while the Eurocord patients had considerable advantage of their KIR-ligand mismatched donor (Table 1). Only 32% of Minnesota patients were transplanted after an ALG/ATG-containing conditioning regimen compared with 81% of all Eurocord patients. It may be possible that the in vivo T-cell depletion of donor and patient secondary to ATG administration contributed to post-transplant expansion of functional NK cells and facilitated alloreactivity in the presence of a KIR-ligand incompatibility. Two UCB units were used in a considerable number of patients by the Minnesota group. They choose to use the dominant engrafting unit of the two UCBs to determine the KIR-ligand (in-) compatibility. However, there is no evidence that the algorithm for donor-versus-recipient NK cell alloreactivity as used for haploidentical stem cell transplantation and single UCBT may be used similarly for double UCBT. Because after double UCBT, three parties may be interacting, a possible role of KIR-ligand incompatibility will be more difficult to investigate. Recently, we studied the effect of KIR-ligand (in-)compatibility in patients with non-malignant disorders transplanted with a single cord blood unit, as NK alloreactivity has been associated with better engraftment and decreased GvHD after haploHSCT [25,26]. With this goal, Eurocord registry analysed 137 patients, mostly children (n ¼ 124) given an UCBT in 47 EBMT centres [46]. A total of 43% of patients had bone-marrow failure syndromes (n ¼ 60). Other diagnoses included metabolic
288
R. Willemze et al. / Best Practice & Research Clinical Haematology 23 (2010) 283–290
Table 1 Patients, disease, transplant characteristics and outcomes by conditioning regimen and KIR incompatibility between Minnesota [44] and Eurocord [45] studies Patient, disease and transplant related factors
Minnesota study [44]
Eurocord study [45]
n Center Median age in years (range)
257 Single 15 years (06–59) (MAC* recipients) 50 years (6–69) (RIC** recipients) 58% 33% 25% 42%
218 57 centers 14 (1–69)
Acute leukemia AML£ ALLþ Other diagnosis Status of the disease Remission Non remission HLA matching 4/6 5/6 KIR-ligand mismatching MAC* RIC** ATG/ALG Single cord blood graft Double cord blood graft Outcomes Acute GVHD (II–IV)
2 years treatment related mortality
2 years relapse rate
3 and 2 years overall survival, respectively
100% 43% 57% 0%
63% 37%
100%
61% 31% 29% 60% 40% 35% 35% 65%
43% 43% 32% 83% 17% 81% 100% 0%
RIC KIR incompatibility (n ¼ 33) 42% KIR compatibility (n ¼ 69) 13% p < 0.001 MAC KIR incompatibility (n ¼ 41) 17% KIR compatibility (n ¼ 114) 17% p ¼ ns RIC KIR incompatibility (n ¼ 33) 27% KIR compatibility (n ¼ 69) 12% p ¼ 0.03 MAC KIR incompatibility (n ¼ 41) 27% KIR compatibility (n ¼ 114) 18% p ¼ ns RIC KIR incompatibility (n ¼ 33) 39% KIR compatibility (n ¼ 69) 47% p ¼ ns MAC KIR incompatibility (n ¼ 41) 18% KIR compatibility (n ¼ 114) 28% p ¼ ns RIC KIR incompatibility (n ¼ 33) 32% KIR compatibility (n ¼ 69) 52% p ¼ 0.03 MAC KIR incompatibility (n ¼ 41) 50% KIR compatibility (n ¼ 114) 57% p ¼ ns
KIR incompatibility (n ¼ 69) 28% KIR compatibility (n ¼ 149) 30% p ¼ ns
KIR incompatibility (n ¼ 69) 25% KIR compatibility (n ¼ 149) 31% p ¼ ns
KIR incompatibility (n ¼ 69) 20% KIR compatibility (n ¼ 149) 37% p ¼ 0.03
KIR incompatibility (n ¼ 69) 57% KIR compatibility (n ¼ 149) 40% p ¼ 0.02
Abbreviations: *MAC ¼ myeloablative conditioning regimen; **RIC ¼ reduced intensity conditioning regimen; £AML ¼ acute myeloid leukemia, þALL ¼ acute lymphoblastic leukemia; ATG/ALG anti-thymocyte or lymphocyte globulin.
disorders (n ¼ 41), severe combined immuno-deficiency (SCID) (n ¼ 32), haemoglobinopathies (n ¼ 2) and auto-immune diseases (n ¼ 2). The frequency of KIR-ligand incompatible (KIRþ) pairs was 23% (n ¼ 31). Median follow-up was 24 months (4–101 months). Median recipient age was 3 years. UCB units were HLA-matched at six of six (n ¼ 38), five of six (n ¼ 62), four of six (n ¼ 34) and three of six (n ¼ 3). Median cell dose infused was 5.6 107 TNC kg1 and 2.4 105 CD34 cells kg1. The conditioning regimen was myeloablative in 48% of cases and reduced intensity in 52% and included ATG in 90% and fludarabine in 42% of the patients. Cumulative incidence of the day-60-neutrophil recovery was 76 3% (median time: 22 days) and it was not associated with KIR-ligand incompatibilities.
R. Willemze et al. / Best Practice & Research Clinical Haematology 23 (2010) 283–290
289
Cumulative incidence of aGvHD was 21 3%; only three patients out of 31 with KIRþ presented aGvHD (II–IV). The estimated probability of 2-year overall survival was 63 4%; it was 64 5% in the group of KIR- and 63 3% in KIRþ donor–patient pairs (p ¼ 0.5). The association of KIR-ligand incompatibility in the HvG direction was also investigated, with the aim to evaluate whether NK-cell alloreactivity in the HvG direction played a role in graft rejection. The frequency of KIR-ligand incompatible pairs in the HvG direction was 22% and it was not associated with neutrophil recovery or OS. We concluded that KIR-ligand incompatibility in the GvH or HvG direction was not associated with outcomes after UCBT for children with non-malignant disease. Conclusion To summarise, the role of NK-cell alloreactivity in allo-SCT transplantation is far from clear. The continuous progress in knowledge in the field of NK cells, their receptors and their ligands warns against early conclusions. Algorithms for KIR-ligand incompatibility used in the past in the choice of a cord blood donor may not be representative for donor-versus-recipient NK cell alloreactivity in the future. Additional studies are needed to draw definite conclusions on the role of KIR-ligand mismatching after UCBT. Conflict of interest statement None to declare. References [1] Falkenburg JHF, Marijt WAF, Heemskerk MH, et al. Minor histocompatibility antigens as target of graft-versus-leukemia reactions. Curr Opin Hematol 2002;9:497–502. [2] Kolb HJ, Schattenberg A, Goldman JM, et al. Graft-versus-leukemia effect of donor lymphocyte transfusions in marrow grafted patients. Blood 1995;86:2041–50. [3] Farag SS, Fehninger TA, Ruggeri L, et al. Natural killer cell receptors: new biology and insights into the graft-versusleukemia effect. Blood 2002;100:1935–47. [4] Moretta L, Moretta A. Killer immunoglobulin-like receptors. Curr Opin Immunol 2004;16:626–33. [5] Parham P. MHC class I molecules and KIRs in human history, health and survival. Nat Rev Immunol 2005;5:201–14. [6] Lanier LL. NK cell recognition. Annu Rev Immunol 2005;23:225–74. [7] Lodoen MB, Lanier LL. Natural killer cells as an initial defense against pathogens. Curr Opin Immunol 2006 Aug;18(4):391–8. [8] Trinchieri G. Biology of natural killer cells. Adv Immunol 1989;47:187–376. [9] Biron CA, Nguyen KB, Pien GC, et al. Natural killer cells in antiviral defense: function and regulation by innate cytokines. Annu Rev Immunol 1999;17:189–220. [10] Fauriat C, Long EO, Ljunggren HG, et al. Regulation of human NK-cell cytokine and chemokine production by target cell recognition. Blood 2010 Mar 18;115(11):2167–76. *[11] Ljunggren HG, Karre K. In search of the “missing self”: MHC molecules and NK cell recognition. Immunol Today 1990;11: 237–44. [12] Karlhofer FM, Ribaudo RK, Yokoyama WM. MHC class I alloantigen specificity of Ly-49þIL-2-activated natural killer cells. Nature 1992;358:66–70. [13] Vivier E, Anfossi N. Inhibitory NK-cell receptors on T cells: witness of the past, actors of the future. Nat Rev Immunol 2004;4:190–8. [14] Biron CA, Brossay L. NK cells and NKT cells in innate defense against viral infections. Curr Opin Immunol 2001;13:458–64. [15] Caligiuri MA. Human natural killer cells. Blood 2008;112:461–9. [16] Marsh SG, Parham P, Dupont B, et al. Killer-cell immunoglobulin-like receptor (KIR) nomenclature report, 2002. Tissue Antigens 2003;62:79–86. [17] Biassoni R, Cantoni C, Falco M, et al. The human leukocyte antigen (HLA)-C-specific “activatory” or “inhibitory” natural killer cell receptors display highly homologous extracellular domains but differ in their transmembrane and intracytoplasmic portions. J Exp Med 1996;183(2):645–50. [18] Foley BA, De Santis D, Van Beelen E, et al. The reactivity of Bw4þ HLA-B and HLA-A alleles with KIR3DL1: implications for patient and donor suitability for haploidentical stem cell transplantation. Blood 2008;112:435–43. [19] Stern M, Ruggeri L, Capanni M, et al. Human leukocyte antigens A23, A24, and A32 but not A25 are ligands for KIR3DL1. Blood 2008;112:708–10. [20] Hansasuta P, Dong T, Thananchai H, et al. Recognition of HLA-A3 and HLA-A11 by KIR3DL2 is peptide-specific. Eur J Immunol 2004;34:1673–9. [21] Middleton D, Gonzelez F. The extensive polymorphism of KIR genes. Immunology 2009;129:8–19. *[22] Leung W, Iyengar R, Triplett B, et al. Comparison of killer Ig-like receptor genotyping and phenotyping for selection of allogeneic blood stem cell donors. J Immunol 2005;174:6540–5. [23] Leung W, Iyengar R, Turner V, et al. Determinants of antileukemia effects of allogeneic NK cells. J Immunol 2004;172:644–50.
290
R. Willemze et al. / Best Practice & Research Clinical Haematology 23 (2010) 283–290
*[24] Ruggeri L, Mancusi A, Capanni M, et al. Donor natural killer cell allorecognition of missing self in haploidentical hematopoieitc transplantation for acute myeloid leukemia: challenging its predictive value. Blood 2007;110:433–40. [25] Ruggeri L, Capanni M, Casucci M, et al. Role of natural killer cell alloreactivity in HLA-mismatched hematopoietic stem cell transplantation. Blood 1999;94:333–9. *[26] Ruggeri L, Capanni M, Urbani E, et al. Effectiveness of donor natural killer alloreactivity in mismatched haematopoietic transplants. Science 2002;295:2097–100. [27] 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;109:5058–61. [28] Giebel S, Locatelli F, Lamparelli T, et al. Survival advantage with KIR ligand incompatibility in hematopoietic stem cell transplantation from unrelated donors. Blood 2003;102:814–9. [29] Beelen DW, Ottinger HD, Ferencik S, et al. Genotypic inhibitory killer immunoglobulin-like receptor ligand incompatibility enhances the long-term antileukemic effect of unmodified allogeneic hematopoietic stem cell transplantation in patients with myeloid leukemias. Blood 2005;105:2594–600. [30] Hsu K, 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;105:4878–84. [31] Davies SM, Ruggieri L, DeFor T, et al. Evaluation of KIR ligand incompatibility in mismatched unrelated donor hematopoietic transplants. Blood 2002;100:3825–7. [32] Farag SS, Bacigalupo A, Eapen M, et al. The effect of KIR ligand incompatibility on the outcome of unrelated donor transplantation: a report from the Center for International Blood and Bone Marrow Transplant Research, the European Blood and Bone Marrow Transplant Registry, and the Dutch registry. Biol Blood Marrow Transplant 2006;12:876–84. [33] Bornhauser M, Schwerdtfeger R, Marin H, et al. Role of KIR ligand incompatibility in hematopoietic stem cell transplantation using unrelated donors. Blood 2004;103:2860–1. [34] Cook MA, Milligan DW, Fegan CD, 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;103:1521–6. [35] Sun JY, Dagis A, Gaidulis L, et al. Detrimental effect of natual killer cell alloreactivity in T-replete hematopoietic stem cell transplantation for leukaemia patients. Biol Blood Marrow Transplant 2007;13:197–205. [36] Schellekens AJ, Rozemuller EH, Petersen EJ, et al. Activating KIRs exert a crucial role on relapse and overall survival after HLA-identical sibling transplantation. Mol Immunol 2008;45:2255–61. [37] Chen C, Busson M, Rocha V, et al. Activating KIR genes are associated with CMV reactivation and survival after non-T cell depleted HLA-identical sibling bone marrow transplantation for malignant disorders. Bone Marrow Transplant 2006;38: 437–44. *[38] Lowe EJ, Turner V, Handgretinger R, et al. T-cell alloreactivity dominates natural killer cell alloreactivity in minimally Tcell-depleted HLA-non-identical paediatric bone marrow transplantation. Brit J Haematol 2003;123:323–6. [39] Gluckman E, Rocha V. History of the clinical use of umbilical cord blood hematopoietic cells. Cytotherapy 2005;7:219–27. *[40] Gluckman E, Rocha V. Donor selection for unrelated cord blood transplants. Curr Opin Immunol 2006;18:565–70. [41] Rocha V, Cornish J, Sievers EL, et al. Comparison of outcomes of unrelated bone marrow and umbilical cord blood transplants in children with acute leukemia. Blood 2001;97(10):2962–71. *[42] Rocha V, Labopin M, Sanz G, et al. Transplants of umbilical-cord blood or bone marrow from unrelated donors in adults with acute leukemia. N Engl J Med 2004;351:2276–85. *[43] Eapen M, Rubinstein P, Zhang MJ, et al. Outcomes of transplantation of unrelated donor umbilical cord blood and bone marrow in children with acute leukaemia: a comparison study. Lancet 2007;369:1947–54. *[44] Brunstein CG, Wagner JE, Weisdorf DJ, et al. Negative effect of KIR alloreactivity in recipients of umbilical cord blood transplant depends on transplantation conditioning intensity. Blood 2009;113:5628–34. *[45] Willemze R, Rodrigues CA, Labopin M, et alEurocord-Netcord and Acute Leukaemia Working Party of the EBMT. KIRligand incompatibility in the graft-versus-host direction improves outcomes after umbilical cord blood transplantation for acute leukemia. Leukemia 2009;23:492–500. [46] Ruggeri A, Purtill D, Michel G, et al. Outcomes of unrelated cord blood transplantation for non-malignant diseases in children are not associated with KIR-ligand matching between donor and recipient: a study on behalf of Eurocord. Blood Nov 2009;114:2301.
Contents lists available at ScienceDirect
Best Practice & Research Clinical Haematology journal homepage: www.elsevier.com/locate/beha
13
Is there an impact of killer cell immunoglobulin-like receptors and KIR-ligand incompatibilities on outcomes after unrelated cord blood stem cell transplantation? Roel Willemze, MD *, Annalisa Ruggeri, MD, Duncan Purtill, MD, Celso Arrais Rodrigues, MD, Eliane Gluckman, MD, Vanderson Rocha on behalf of Eurocord and of the European Group of Blood and Marrow Transplantation Department of Hematology, Leiden University Medical Center, Albinusdreef 2, Leiden, the Netherlands
Keywords: KIR KIR-ligand incompatibility umbilical cord blood transplantation
Donor killer cell immunoglobulin-like receptor (KIR) ligand incompatibility in the graft-versus-host direction is associated with decreased relapse incidence and improved disease-free survival after haploidentical and human leucocyte antigen (HLA)mismatched unrelated, haematopoietic stem cell transplantation. However, review of all published studies of allogeneic HLAmatched or mismatched stem cell transplantation shows that the results on the relationship between donor–recipient KIR(-ligand) (in)compatibility and outcomes are highly variable, ranging from highly beneficial to detrimental. Reasons for these differences may include the methodology to determine KIR(-ligand) incompatibility, the disease distribution and the transplant protocol or donor type. Two retrospective studies on the effects of KIR-ligand incompatibility in unrelated cord blood transplantation (UCBT) for haematological malignancies have resulted in conflicting results. The Eurocord study showed a favourable effect of KIR-ligand mismatching on relapse incidence and leukaemia-free survival, whereas the Minneapolis study showed no effect on these end points and a detrimental effect on incidence of graft-versus-host disease (GvHD). In patients with non-malignant disorders, KIRligand (in)compatibility between donor and recipient was not associated with outcomes in a recent Eurocord analysis. Therefore,
* Corresponding author. E-mail address: [email protected] (R. Willemze). 1521-6926/$ – see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.beha.2010.05.005
284
R. Willemze et al. / Best Practice & Research Clinical Haematology 23 (2010) 283–290
the role of natural killer (NK) cell alloreactivity in UCBT is far from clear. It is too early to use a donor–recipient KIR(-ligand) algorithm for selection of a cord blood donor. Ó 2010 Elsevier Ltd. All rights reserved.
Allogeneic haematopoietic stem cell transplantation Allogeneic human leucocyte antigen-(HLA) matched or mismatched haemato- poietic stem cell transplantation (allo-SCT) has been applied for decades as a method of stem cell reconstitution in patients who have received myeloablative chemotherapy and irradiation for the treatment of haematological diseases. Usually, the origin of the stem cell graft is either bone marrow or mobilised stem cells from peripheral blood. Since 15 years, umbilical cord blood units from unrelated donors constitute another well-established source of stem cells for haematopoietic stem cell transplantation. Initially, the only purpose of stem cell transplantation was to replace the irreversibly damaged haematopoietic stem cell compartment after chemotherapy and radiotherapy at the maximum doses that could be tolerated by non-haematopoietic tissues. However, in contrast to autologous stem cell transplantation, allo-SCT is accompanied by extra immunological benefits, such as antitumour response, and complications such as graft-versus-host disease (GvHD), immune incompetence or graft rejection. It became apparent that donor-derived T-cells play an important role in several of these complications, as well as in the incidence of recurrence of the original haematological disease by recognising differences in minor and major histocompatibility antigens between the donor and the recipient [1,2]. More recently, a novel immunological mechanism with clinical implications after allogeneic stem cell transplantation that relies on interactions between donor natural killer (NK) cells and recipient cells, was demonstrated in the setting of haploidentical family donor SCT [3–5].
Natural Killer (NK) cells NK cells are large granular lymphocytes originating in the bone marrow from CD34þ cells. Owing to their early production of cytokines and chemokines (interferon(IFN)-g, tumour necrosis factor (TNF)a and granulocyte macrophage-colony stimulating factor (GM-CSF) and their ability to lyse target cells without prior sensitisation, NK cells (10–15% of all lymphocytes; CD56þ, CD3) are, as innate immune lymphocytes, critical to host defence against invading infectious pathogens and malignant transformation [6]. NK cells can be distinguished from other lymphocytes by the expression of the surface markers CD56 and CD16 and the absence of CD3. Based on the cell surface density of CD56 and CD16, human NK cells can be divided into two subsets, the regulatory and the cytotoxic NK cells [7]. The immunoregulatory CD56bright, CD16dim or negative cells (10% of NK cells) are large agranular cells, that express intermediate and high-affinity immunoglobulin-like (IL)-2, and high CD94/NKG2 and low killer cell immunoglobulin-like receptor (KIR) receptors, and that secrete, upon exogenous stimulation with pro-inflammatory cytokines, a greater amount of cytokines and chemokines than CD56dim NK cells [8,9]. The cytotoxic CD56dim, CD16 positive cells (90% of NK cells) are cells with the cytoplasm containing cytolytic granules that express intermediate affinity for IL-2, high CD94/NKG2 and high KIR receptors and are the effectors of natural cell lysis or antibody-dependent cellular cytotoxicity (ADCC) [8,9]. Recently, it was shown that in response to target recognition, these cells, rather than CD56bright NK cells, are primarily responsible for production of cytokines and chemokines [10]. The mechanism behind NK cell cytolytic activity was first proposed by Ljunggren and Karre[11] in 1990 as the ‘missing self’ hypothesis. They suggested that NK cells kill target cells when they lack or have a low expression of ‘self’ HLA class I molecules. The molecular proof came in 1992 with the identification of inhibitory NK-cell receptors [12]. Since then, a wide range of both inhibitory and activating receptors that regulate NK-cell function has been identified [13–15].
R. Willemze et al. / Best Practice & Research Clinical Haematology 23 (2010) 283–290
285
Killer cell immunoglobulin-like receptors The cytolytic activity of human NK cells is modulated by the interaction of two types of structurally different, inhibitory and activating, major histocompatibility complex (MHC) Class 1 specific receptor superfamilies, the CD94–NKG2 family and the killer cell immunoglobulin-like receptors (KIR) family, with MHC class 1 antigens expressed by host cells. The CD94–NKG2 family comprises the C-type lectinslike receptors, which form heterodimers of the invariant CD94 chain and different members of the NKG2 family. These receptors are encoded on chromosome 12. The inhibitory CD94–NKG2A receptors fulfil an important function in the prevention of NK cell autoreactivity. They recognise the non-classical HLA-E class Ib molecules. The immunoglobulin-like receptor family includes the KIRs. KIRs are encoded on chromosome 19 and are expressed on NK cells and a subset of T-cells. They play an important role in the prevention of autoreactivity and the elimination of viral-infected and tumour cells [13–15]. The KIRs are transmembrane receptors encoded by 17 genes. The nomenclature of these genes is based on three structural criteria: the number of extracellular immunoglobulin domains that varies between two (2D) or three (3D) domains, the presence of a short (S) or long (L) intracellular domain (cytoplasmic tail) and the sequence similarity [16]. These criteria define the functionally relevant KIR2DL1-5, KIR3DL1-3, KIR2DS1-5 and KIR3DS1. The structural characteristics correlate with function. The length of the cytoplasmic tail conveys the type of signal that is generated upon ligand binding. The KIRs with a long cytoplasmic tail inhibit NK cell reactivity by recruitment of the phosphatase Src homology 2 domain-containing phosphatase 1 (SHP-1) to the receptors’ phosphorylated cytoplasmic immunoreceptor thyrosine-based inhibition motifs (ITIMs). KIRs with a short cytoplasmic tail lack these ITIMs but have a positively charged lysine residue in the transmembrane domain, which associates with the DAP12 signalling molecule and thereby induces NK reactivity [15]. KIR ligands The HLA class I molecules act as ligands for some of the KIR genes. The majority of KIRs is specific for HLA-C, but HLA-A, -B and -G have also been described as ligands for some KIRs. KIRs do not recognise each of the allelic variants of the HLA-C molecule, but are restricted to two known epitope groups of HLA-C. HLA-C group 1 (C1) covers the alleles with an asparagine at position 80, while the group 2 (C2) alleles have a lysine at that position. Each of both epitope groups are recognised by inhibitory and activating KIRs [17]. The inhibitory KIR2DL2/KIR2DL3 receptors and the activating KIR2DS2 recognise HLA-C group 1 epitopes, and the inhibitory KIR2DL1 and the activating KIR2DS1 receptor recognise HLA-C group 2 epitopes. KIR3DL1 and KIR3DS1 are inhibitory and activating receptors, respectively, for HLA-B allotypes with Bw4 motifs at positions 77–83. Recently, the specificity of KIR3DL1 has been found to be restricted to all Bw4 alleles (except for 1301 and 1302) and to some HLA-A alleles such as 2301, 2402 and 3201[18,19]; KIR2DL4 recognises HLA-G. KIR3DL2 has as its ligand HLA-A3 and HLA-A11 allele families, but only when certain virally derived peptides are loaded [20]. Donor-versus-recipient NK cell alloreactivity in allo-SCT Donor NK cells generate alloreactions against recipient cells through their KIR receptors when they are mismatched with the recipient for the HLA-C groups and/or the HLA-Bw4 group. Furthermore, HLA-A3 and HLA-A11 function as ligands for KIR3DL2 presumably only when binding EBV reactivation in vitro.Therefore, there is some controversy on the use of HLA-A3 and HLA-A11 as KIR ligand in the clinical setting [20]. The genes encoding inhibitory KIR are nearly always present in populations at frequencies greater than 90%. These genes are not necessarily expressed as corresponding receptor by the NK cell and if expressed, surface expression levels and strength of inhibitory signal may vary considerably [21–23]. Functional analysis of NK clones against HLA-C group 1 or group 2, or HLA-Bw4 tested with cells from mismatched recipients showed alloreactivity in virtual all HLA-C mismatched cases tested, but only in a number of HLA-Bw4 mismatched combinations. No donor alloreactive NK clones could be detected in donor–recipient combinations without KIR-ligand mismatches in the GvH direction, indicating that KIR-ligand mismatching may be a prerequisite for NK cell alloreactivity [24].
286
R. Willemze et al. / Best Practice & Research Clinical Haematology 23 (2010) 283–290
Donor-versus-recipient NK cell alloreactivity and allogeneic blood or bone-marrow-derived stem cell transplantation The role of KIRs and KIR-ligand in the context of allogeneic blood or bone marrow stem cell transplantation has been investigated over the past 10 years. In transplants that are KIR-ligand mismatched in the GvH direction, donor NK cells expressing inhibitory KIRs, which do not recognise ligand(s) on recipient targets, are released from HLA inhibition and mediate allo-responses leading to clinically significant effects. Clinical proof for these NK cell-induced effects were reported by Ruggeri et al. [25,26], who, using haplo-identical family donor SCT, has shown that KIR or KIR-ligand incompatibility in the GvH direction was associated with a decreased incidence of graft failure, acute GvHD and relapse and an improved survival of patients with acute myeloid leukaemia (AML). They explain their clinical results by findings from mouse and human experiments in which donor-versus-recipient alloreactive NK cells eliminate residual leukaemic cells, ablate host-type dendritic cells responsible for triggering GvHD, and attack residual host lympho–haematopoietic cells including T-cells that are responsible for graft rejection [26]. Different algorithms have been used to investigate donor-versus-recipient NK-cell alloreactivity after allo-SCT. The most reliable method to determine NK-cell alloreactivity is probably by donor KIR clones against KIR-ligands on recipient cells or by donor KIR genotyping. However, since data on KIR genotype are usually not readily available, the most frequently applied algorithm is the usage of the presence of KIR-ligand compatibility or incompatibility between the donor and the recipient. These algorithms have applied to investigate the clinical relevance of donor–recipient NK-cell alloreactivity in a number of series of HLA-matched or mismatched related- or unrelated stem cell transplantations. The results of these studies with respect to the outcome of allogeneic SCT were highly variable, ranging from highly beneficial to detrimental [27–37]. These differences in outcomes may be partly due to many differences between the studies. First, as already mentioned, the methods to determine donor-versus-recipient NK-cell alloreactivity varied. Furthermore, some authors included only HLA-C and HLA-B, others also included HLA-A KIR-ligands; and in some studies inhibitory as well as activating KIR receptors were included in the analysis [36,37]. Second, inclusion of patients of all ages with various haematological malignancies in different stages of the disease, transplanted with stem cells from HLA-matched as well as mismatched, family or unrelated donors after various kinds of conditioning regimens may have influenced outcomes with respect to NK-cell reactivity. Third, the number of donor or recipient T-cells involved in the transplantation procedure differed among the studies. T- and NK-cells of the donor have been shown to play a role in HLA-matched as well as mismatched haematopoietic stem cell transplantation (HSCT). To what extent both alloreactivities influence each other in the clinical setting, is not extensively studied. Lowe et al. [38] suggested that Tcell alloreactivity may mask NK-cell alloreactivity in minimally T-cell-depleted grafts of HLA-nonidentical donors. The published studies fail to show a clear pattern with respect to T-cell depletion and the effects of NK-cell alloreactivity. Extensive T-cell depletion by CD34þcell selection and intravenous anti-thymocyte or anti-lymphocyte globulin (ATG/ALG) in the conditioning regimen has only been performed in the haploidentical setting [24–26], a schedule that is associated with rapid recovery of NK cells and slow T-cell recovery, but with the strongest effects of donor-versus-recipient NK reactivity. In the remaining studies where ATG in the conditioning regimen may have been responsible for some in vivo T-cell depletion, conflicting results are reported. Donor-versus-recipient NK cell alloreactivity and unrelated cord blood stem cell transplantation for patients with malignant and non-malignant diseases Umbilical cord blood stem cell transplantation represents an increasingly popular method of allogeneic HSCT. The main advantages are rapid availability and the possibility to perform HLA-mismatched transplantation, due to a relatively low risk of severe GvHD. This increases considerably the probability of finding a suitable donor, since one or two HLA disparities (on the basis of antigen level HLA-A and HLA-B and allele level HLA-DRB1 typing) between donor and recipient lead to similar outcomes and only more than two HLA disparities are associated with diminished neutrophil recovery, increased non-relapse mortality and decreased survival chances [39,40].
R. Willemze et al. / Best Practice & Research Clinical Haematology 23 (2010) 283–290
287
Moreover, outcomes after unrelated cord blood transplantation (UCBT) are comparable with those after unrelated bone marrow transplantation in children and adults with acute leukaemia [41–43]. Only two retrospective studies have been currently published investigating the possible effects of donor-versus-recipient NK cell alloreactivity after UCBT [44,45]. The Eurocord–Netcord and Acute Leukemia Working Party of the EBMT [45] assessed, using the Eurocord Registry, NK cell alloreactivity by means of KIR-ligand (in)compatibility in the GvH direction in 218 patients with acute leukaemia (AML n ¼ 94, acute lymphoblastic leukaemia (ALL) n ¼ 124) in complete remission, who had received a single unit UCBT from a KIR-ligand compatible or KIR-ligandincompatible donor. Patients and donors were categorised to their degree of KIR-ligand compatibility by determining whether or not they expressed HLA-C group 1 or 2, HLA-Bw4 or HLA-A3/-A11. Sixtynine donor–patient combinations were KIR-ligand mismatched in the GvH direction and 149 were matched. They found that, with a median follow-up time of only 14 months, KIR-ligand incompatibility in the GvH direction between donor and recipient was associated with a significantly reduced incidence of relapse (2 years cumulative incidence 20% vs. 37%, p ¼ 0.03), improved leukaemia-free survival (55% vs. 31%, p ¼ 0.005) and overall survival (OS) (57% vs. 40%, p ¼ 0.02). No effect was found on engraftment pattern, acute or chronic GvHD and non-relapse mortality. These results on relapse incidence, leukaemia-free survival and OS were even more evident for AML transplant recipients [45]. A recently performed update of the data of this series of patients by Annalisa Ruggeri did not show, after a median follow up of 34 months, any significant change in the outcomes. By contrast, Brunstein et al. [44] reported a lack of, or even detrimental, effects of KIR-ligand alloreactivity on acute graft-versus-host disease (aGvHD), relapse incidence (RI) and OS, in 257 recipients of a single (n ¼ 91) or a double (n ¼ 166) unit UCBT following myeloablative (n ¼ 155) or reduced intensity (n ¼ 102) conditioning regimens. After myeloablative conditioning, KIR-ligand incompatibility had no effect on grade II–IV acute GvHD, transplant-related mortality, relapse and survival. By contrast, following reduced intensity conditioning, KIR-ligand incompatibility resulted in higher rates of acute GvHD (42% vs. 13%, p < 0.01) and treatment related mortality (27% vs. 12%, p ¼ 0.03) with inferior survival (32% vs. 52%, p ¼ 0.03). There were no apparent methodological differences between the two analyses with respect to the algorithm of KIR-ligand incompatibility, since the Minnesota group included, just as the Eurocord group, HLA-C groups 1 and 2, HLA-Bw4 as well as the controversial HLA-A3/11 as KIR-ligand for their calculations. In addition, the frequency of KIR-ligand incompatibility in the GvH direction was equal in the two studies (29% vs. 32%, respectively). There are a number of differences with respect to median age, diagnoses, HLA-matching, conditioning regimens, use of ATG and the number of umbilical cord blood (UCB) units used. Diagnosis as explanation can be ruled out since comparing the outcomes for AML only as reported by the Minnesota (n ¼ 88; 62 KIR-ligand matched, 26 mismatched) and by the Eurocord group (n ¼ 94; 68 KIR-ligand matched, 26 mismatched) showed that the Minneapolis group did not see significant effects of KIR-ligand incompatibility, while the Eurocord patients had considerable advantage of their KIR-ligand mismatched donor (Table 1). Only 32% of Minnesota patients were transplanted after an ALG/ATG-containing conditioning regimen compared with 81% of all Eurocord patients. It may be possible that the in vivo T-cell depletion of donor and patient secondary to ATG administration contributed to post-transplant expansion of functional NK cells and facilitated alloreactivity in the presence of a KIR-ligand incompatibility. Two UCB units were used in a considerable number of patients by the Minnesota group. They choose to use the dominant engrafting unit of the two UCBs to determine the KIR-ligand (in-) compatibility. However, there is no evidence that the algorithm for donor-versus-recipient NK cell alloreactivity as used for haploidentical stem cell transplantation and single UCBT may be used similarly for double UCBT. Because after double UCBT, three parties may be interacting, a possible role of KIR-ligand incompatibility will be more difficult to investigate. Recently, we studied the effect of KIR-ligand (in-)compatibility in patients with non-malignant disorders transplanted with a single cord blood unit, as NK alloreactivity has been associated with better engraftment and decreased GvHD after haploHSCT [25,26]. With this goal, Eurocord registry analysed 137 patients, mostly children (n ¼ 124) given an UCBT in 47 EBMT centres [46]. A total of 43% of patients had bone-marrow failure syndromes (n ¼ 60). Other diagnoses included metabolic
288
R. Willemze et al. / Best Practice & Research Clinical Haematology 23 (2010) 283–290
Table 1 Patients, disease, transplant characteristics and outcomes by conditioning regimen and KIR incompatibility between Minnesota [44] and Eurocord [45] studies Patient, disease and transplant related factors
Minnesota study [44]
Eurocord study [45]
n Center Median age in years (range)
257 Single 15 years (06–59) (MAC* recipients) 50 years (6–69) (RIC** recipients) 58% 33% 25% 42%
218 57 centers 14 (1–69)
Acute leukemia AML£ ALLþ Other diagnosis Status of the disease Remission Non remission HLA matching 4/6 5/6 KIR-ligand mismatching MAC* RIC** ATG/ALG Single cord blood graft Double cord blood graft Outcomes Acute GVHD (II–IV)
2 years treatment related mortality
2 years relapse rate
3 and 2 years overall survival, respectively
100% 43% 57% 0%
63% 37%
100%
61% 31% 29% 60% 40% 35% 35% 65%
43% 43% 32% 83% 17% 81% 100% 0%
RIC KIR incompatibility (n ¼ 33) 42% KIR compatibility (n ¼ 69) 13% p < 0.001 MAC KIR incompatibility (n ¼ 41) 17% KIR compatibility (n ¼ 114) 17% p ¼ ns RIC KIR incompatibility (n ¼ 33) 27% KIR compatibility (n ¼ 69) 12% p ¼ 0.03 MAC KIR incompatibility (n ¼ 41) 27% KIR compatibility (n ¼ 114) 18% p ¼ ns RIC KIR incompatibility (n ¼ 33) 39% KIR compatibility (n ¼ 69) 47% p ¼ ns MAC KIR incompatibility (n ¼ 41) 18% KIR compatibility (n ¼ 114) 28% p ¼ ns RIC KIR incompatibility (n ¼ 33) 32% KIR compatibility (n ¼ 69) 52% p ¼ 0.03 MAC KIR incompatibility (n ¼ 41) 50% KIR compatibility (n ¼ 114) 57% p ¼ ns
KIR incompatibility (n ¼ 69) 28% KIR compatibility (n ¼ 149) 30% p ¼ ns
KIR incompatibility (n ¼ 69) 25% KIR compatibility (n ¼ 149) 31% p ¼ ns
KIR incompatibility (n ¼ 69) 20% KIR compatibility (n ¼ 149) 37% p ¼ 0.03
KIR incompatibility (n ¼ 69) 57% KIR compatibility (n ¼ 149) 40% p ¼ 0.02
Abbreviations: *MAC ¼ myeloablative conditioning regimen; **RIC ¼ reduced intensity conditioning regimen; £AML ¼ acute myeloid leukemia, þALL ¼ acute lymphoblastic leukemia; ATG/ALG anti-thymocyte or lymphocyte globulin.
disorders (n ¼ 41), severe combined immuno-deficiency (SCID) (n ¼ 32), haemoglobinopathies (n ¼ 2) and auto-immune diseases (n ¼ 2). The frequency of KIR-ligand incompatible (KIRþ) pairs was 23% (n ¼ 31). Median follow-up was 24 months (4–101 months). Median recipient age was 3 years. UCB units were HLA-matched at six of six (n ¼ 38), five of six (n ¼ 62), four of six (n ¼ 34) and three of six (n ¼ 3). Median cell dose infused was 5.6 107 TNC kg1 and 2.4 105 CD34 cells kg1. The conditioning regimen was myeloablative in 48% of cases and reduced intensity in 52% and included ATG in 90% and fludarabine in 42% of the patients. Cumulative incidence of the day-60-neutrophil recovery was 76 3% (median time: 22 days) and it was not associated with KIR-ligand incompatibilities.
R. Willemze et al. / Best Practice & Research Clinical Haematology 23 (2010) 283–290
289
Cumulative incidence of aGvHD was 21 3%; only three patients out of 31 with KIRþ presented aGvHD (II–IV). The estimated probability of 2-year overall survival was 63 4%; it was 64 5% in the group of KIR- and 63 3% in KIRþ donor–patient pairs (p ¼ 0.5). The association of KIR-ligand incompatibility in the HvG direction was also investigated, with the aim to evaluate whether NK-cell alloreactivity in the HvG direction played a role in graft rejection. The frequency of KIR-ligand incompatible pairs in the HvG direction was 22% and it was not associated with neutrophil recovery or OS. We concluded that KIR-ligand incompatibility in the GvH or HvG direction was not associated with outcomes after UCBT for children with non-malignant disease. Conclusion To summarise, the role of NK-cell alloreactivity in allo-SCT transplantation is far from clear. The continuous progress in knowledge in the field of NK cells, their receptors and their ligands warns against early conclusions. Algorithms for KIR-ligand incompatibility used in the past in the choice of a cord blood donor may not be representative for donor-versus-recipient NK cell alloreactivity in the future. Additional studies are needed to draw definite conclusions on the role of KIR-ligand mismatching after UCBT. Conflict of interest statement None to declare. References [1] Falkenburg JHF, Marijt WAF, Heemskerk MH, et al. Minor histocompatibility antigens as target of graft-versus-leukemia reactions. Curr Opin Hematol 2002;9:497–502. [2] Kolb HJ, Schattenberg A, Goldman JM, et al. Graft-versus-leukemia effect of donor lymphocyte transfusions in marrow grafted patients. Blood 1995;86:2041–50. [3] Farag SS, Fehninger TA, Ruggeri L, et al. Natural killer cell receptors: new biology and insights into the graft-versusleukemia effect. Blood 2002;100:1935–47. [4] Moretta L, Moretta A. Killer immunoglobulin-like receptors. Curr Opin Immunol 2004;16:626–33. [5] Parham P. MHC class I molecules and KIRs in human history, health and survival. Nat Rev Immunol 2005;5:201–14. [6] Lanier LL. NK cell recognition. Annu Rev Immunol 2005;23:225–74. [7] Lodoen MB, Lanier LL. Natural killer cells as an initial defense against pathogens. Curr Opin Immunol 2006 Aug;18(4):391–8. [8] Trinchieri G. Biology of natural killer cells. Adv Immunol 1989;47:187–376. [9] Biron CA, Nguyen KB, Pien GC, et al. Natural killer cells in antiviral defense: function and regulation by innate cytokines. Annu Rev Immunol 1999;17:189–220. [10] Fauriat C, Long EO, Ljunggren HG, et al. Regulation of human NK-cell cytokine and chemokine production by target cell recognition. Blood 2010 Mar 18;115(11):2167–76. *[11] Ljunggren HG, Karre K. In search of the “missing self”: MHC molecules and NK cell recognition. Immunol Today 1990;11: 237–44. [12] Karlhofer FM, Ribaudo RK, Yokoyama WM. MHC class I alloantigen specificity of Ly-49þIL-2-activated natural killer cells. Nature 1992;358:66–70. [13] Vivier E, Anfossi N. Inhibitory NK-cell receptors on T cells: witness of the past, actors of the future. Nat Rev Immunol 2004;4:190–8. [14] Biron CA, Brossay L. NK cells and NKT cells in innate defense against viral infections. Curr Opin Immunol 2001;13:458–64. [15] Caligiuri MA. Human natural killer cells. Blood 2008;112:461–9. [16] Marsh SG, Parham P, Dupont B, et al. Killer-cell immunoglobulin-like receptor (KIR) nomenclature report, 2002. Tissue Antigens 2003;62:79–86. [17] Biassoni R, Cantoni C, Falco M, et al. The human leukocyte antigen (HLA)-C-specific “activatory” or “inhibitory” natural killer cell receptors display highly homologous extracellular domains but differ in their transmembrane and intracytoplasmic portions. J Exp Med 1996;183(2):645–50. [18] Foley BA, De Santis D, Van Beelen E, et al. The reactivity of Bw4þ HLA-B and HLA-A alleles with KIR3DL1: implications for patient and donor suitability for haploidentical stem cell transplantation. Blood 2008;112:435–43. [19] Stern M, Ruggeri L, Capanni M, et al. Human leukocyte antigens A23, A24, and A32 but not A25 are ligands for KIR3DL1. Blood 2008;112:708–10. [20] Hansasuta P, Dong T, Thananchai H, et al. Recognition of HLA-A3 and HLA-A11 by KIR3DL2 is peptide-specific. Eur J Immunol 2004;34:1673–9. [21] Middleton D, Gonzelez F. The extensive polymorphism of KIR genes. Immunology 2009;129:8–19. *[22] Leung W, Iyengar R, Triplett B, et al. Comparison of killer Ig-like receptor genotyping and phenotyping for selection of allogeneic blood stem cell donors. J Immunol 2005;174:6540–5. [23] Leung W, Iyengar R, Turner V, et al. Determinants of antileukemia effects of allogeneic NK cells. J Immunol 2004;172:644–50.
290
R. Willemze et al. / Best Practice & Research Clinical Haematology 23 (2010) 283–290
*[24] Ruggeri L, Mancusi A, Capanni M, et al. Donor natural killer cell allorecognition of missing self in haploidentical hematopoieitc transplantation for acute myeloid leukemia: challenging its predictive value. Blood 2007;110:433–40. [25] Ruggeri L, Capanni M, Casucci M, et al. Role of natural killer cell alloreactivity in HLA-mismatched hematopoietic stem cell transplantation. Blood 1999;94:333–9. *[26] Ruggeri L, Capanni M, Urbani E, et al. Effectiveness of donor natural killer alloreactivity in mismatched haematopoietic transplants. Science 2002;295:2097–100. [27] 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;109:5058–61. [28] Giebel S, Locatelli F, Lamparelli T, et al. Survival advantage with KIR ligand incompatibility in hematopoietic stem cell transplantation from unrelated donors. Blood 2003;102:814–9. [29] Beelen DW, Ottinger HD, Ferencik S, et al. Genotypic inhibitory killer immunoglobulin-like receptor ligand incompatibility enhances the long-term antileukemic effect of unmodified allogeneic hematopoietic stem cell transplantation in patients with myeloid leukemias. Blood 2005;105:2594–600. [30] Hsu K, 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;105:4878–84. [31] Davies SM, Ruggieri L, DeFor T, et al. Evaluation of KIR ligand incompatibility in mismatched unrelated donor hematopoietic transplants. Blood 2002;100:3825–7. [32] Farag SS, Bacigalupo A, Eapen M, et al. The effect of KIR ligand incompatibility on the outcome of unrelated donor transplantation: a report from the Center for International Blood and Bone Marrow Transplant Research, the European Blood and Bone Marrow Transplant Registry, and the Dutch registry. Biol Blood Marrow Transplant 2006;12:876–84. [33] Bornhauser M, Schwerdtfeger R, Marin H, et al. Role of KIR ligand incompatibility in hematopoietic stem cell transplantation using unrelated donors. Blood 2004;103:2860–1. [34] Cook MA, Milligan DW, Fegan CD, 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;103:1521–6. [35] Sun JY, Dagis A, Gaidulis L, et al. Detrimental effect of natual killer cell alloreactivity in T-replete hematopoietic stem cell transplantation for leukaemia patients. Biol Blood Marrow Transplant 2007;13:197–205. [36] Schellekens AJ, Rozemuller EH, Petersen EJ, et al. Activating KIRs exert a crucial role on relapse and overall survival after HLA-identical sibling transplantation. Mol Immunol 2008;45:2255–61. [37] Chen C, Busson M, Rocha V, et al. Activating KIR genes are associated with CMV reactivation and survival after non-T cell depleted HLA-identical sibling bone marrow transplantation for malignant disorders. Bone Marrow Transplant 2006;38: 437–44. *[38] Lowe EJ, Turner V, Handgretinger R, et al. T-cell alloreactivity dominates natural killer cell alloreactivity in minimally Tcell-depleted HLA-non-identical paediatric bone marrow transplantation. Brit J Haematol 2003;123:323–6. [39] Gluckman E, Rocha V. History of the clinical use of umbilical cord blood hematopoietic cells. Cytotherapy 2005;7:219–27. *[40] Gluckman E, Rocha V. Donor selection for unrelated cord blood transplants. Curr Opin Immunol 2006;18:565–70. [41] Rocha V, Cornish J, Sievers EL, et al. Comparison of outcomes of unrelated bone marrow and umbilical cord blood transplants in children with acute leukemia. Blood 2001;97(10):2962–71. *[42] Rocha V, Labopin M, Sanz G, et al. Transplants of umbilical-cord blood or bone marrow from unrelated donors in adults with acute leukemia. N Engl J Med 2004;351:2276–85. *[43] Eapen M, Rubinstein P, Zhang MJ, et al. Outcomes of transplantation of unrelated donor umbilical cord blood and bone marrow in children with acute leukaemia: a comparison study. Lancet 2007;369:1947–54. *[44] Brunstein CG, Wagner JE, Weisdorf DJ, et al. Negative effect of KIR alloreactivity in recipients of umbilical cord blood transplant depends on transplantation conditioning intensity. Blood 2009;113:5628–34. *[45] Willemze R, Rodrigues CA, Labopin M, et alEurocord-Netcord and Acute Leukaemia Working Party of the EBMT. KIRligand incompatibility in the graft-versus-host direction improves outcomes after umbilical cord blood transplantation for acute leukemia. Leukemia 2009;23:492–500. [46] Ruggeri A, Purtill D, Michel G, et al. Outcomes of unrelated cord blood transplantation for non-malignant diseases in children are not associated with KIR-ligand matching between donor and recipient: a study on behalf of Eurocord. Blood Nov 2009;114:2301.