Immunology

KIR2DS1 genotype predicts for complete cytogenetic response and survival in newly diagnosed chronic myeloid leukemia patients treated with imatinib

Abstract

Natural killer (NK) cells are expanded in chronic myeloid leukemia (CML) patients on tyrosine kinase inhibitors (TKI) and exert cytotoxicity. The inherited repertoire of killer immunoglobulin-like receptors (KIR) may influence response to TKI. We investigated the impact of KIR-genotype on outcome in 166 chronic phase CML patients on first-line imatinib treatment. We validated our findings in an independent patient group. On multivariate analysis, KIR2DS1 genotype (RR=1.51, P=0.03) and Sokal risk score (low-risk RR=1, intermediate-risk RR=1.53, P=0.04, high-risk RR=1.69, P=0.034) were the only independent predictors for failure to achieve complete cytogenetic response (CCyR). Furthermore, KIR2DS1 was the only factor predicting shorter progression-free (PFS) (RR=3.1, P=0.03) and overall survival (OS) (RR=2.6, P=0.04). The association between KIR2DS1 and CCyR, PFS and OS was validated by KIR genotyping in 174 CML patients on first-line imatinib in the UK multi-center SPIRIT-1 trial; in this cohort, KIR2DS1(+) patients had significantly lower 2-year probabilities of achieving CCyR (76.9 vs 87.9%, P=0.003), PFS (85.3 vs 98.1%, P=0.007) and OS (94.4 vs 100%, P=0.015) than KIR2DS1(−) patients. The impact of KIR2DS1 on CCyR was greatest when the ligand for the corresponding inhibitory receptor, KIR2DL1, was absent (P=0.00006). Our data suggest a novel role for KIR-HLA immunogenetics in CML patients on TKI.

Introduction

Tyrosine kinase inhibitors (TKIs) have revolutionized the treatment of chronic myeloid leukemia (CML), but despite their proven efficacy to induce hematologic and cytogenetic responses, the majority of patients have persisting disease as assessed by quantitative real-time PCR for BCR-ABL1 transcripts.1, 2 This may be due to survival of quiescent leukemic stem cells, which have been shown to be resistant to TKIs. Currently, the only treatment modality which offers the possibility of long-term disease-free survival remains allogeneic hematopoietic stem cell transplantation (allo-SCT), the ‘curative’ effect of which is mediated in large through an allo-immune graft-vs-leukemia effect.3

Natural killer (NK) cells are an important component of the innate immune system, providing first-line defense against altered tissues, notably virally infected cells and tumors. The physiological functions of NK cells, including cytotoxicity and cytokine release, are determined by a balance between signals received from inhibitory and activating cell surface receptors. These receptors include the killer immunoglobulin-like receptors (KIRs), which are specific for allotypic determinants shared by different HLA-class I molecules (referred to as KIR ligands).4, 5, 6

The role of NK cells in the graft-vs-leukemia response has been shown in haplo-identical allo-SCT, and in some studies, in both the matched related and unrelated donor settings7, 8, 9, 10 This, together with in-vitro evidence that NK cells are capable of killing CML progenitor cells and exerting cytotoxic activity against primitive quiescent CD34+ Philadelphia (Ph) positive cells, suggest that NK cells may also have a central role in the outcome of patients with CML.11, 12 Moreover, a recent study has reported the presence of clonal, BCR-ABL1-negative NK and T-lymphocyte populations in patients with CML, which expand further on imatinib and dasatinib.13 Data on the impact of imatinib on NK function have been conflicting, ranging from impaired NK mediated anti-leukemia responses to enhanced anti-tumor NK responses.14, 15

We demonstrated an impact of KIR genotype on the response and outcome of patients with CML treated with imatinib as first-line therapy, suggesting novel immunological interactions between NK and CML cells of clinical relevance.

Patients and methods

Patient characteristics and treatment

A total of 166 consecutive adult patients with BCR-ABL1-positive CML in chronic phase (CP) who received imatinib as first-line therapy at Imperial College NHS Healthcare Trust were included as the ‘discovery set’ in the study. Patient characteristics are detailed in Table 1. The median follow-up was 36 months and 75% of the patients were followed for a minimum of 24 months. No patients were lost to follow-up. Patients were commenced on imatinib (Gleevec, Novartis, Basel, Switzerland) within 6 months of diagnosis with no previous treatment for leukemia other than hydroxycarbamide for initial cytoreduction. All patients gave written informed consent for the use of their data for this analysis. CP was defined by conventional criteria.16, 17 Patients received imatinib 400 mg daily orally and the dose was adjusted according to tolerance and response.18, 19 The dose of imatinib was reduced in the presence of grades III–IV toxicity20 with the aim of maintaining imatinib at or above 300 mg/day. The criteria for changing to a second generation tyrosine kinase inhibitor (2G-TKI) were those recommended by the European LeukemiaNet.21 Thirty-eight patients discontinued imatinib therapy and received either a 2G-TKI (n=22) (dasatinib or nilotinib) or allo-SCT (n=16).

Table 1 Characteristics of patients with chronic myeloid leukemia who received 400 mg Imatinib mesylate as first-line treatment (discovery sample)

Bone marrow morphology and cytogenetics were assessed at diagnosis, and then at three-monthly intervals until patients achieved complete cytogenetic response (CCyR). Thereafter, patients were monitored by real-time PCR for BCR-ABL1 transcripts and annual bone marrow examination. CCyR was defined by the failure to detect any Ph chromosome-positive metaphases in a bone-marrow examination with a minimum of 20 metaphases examined.

Validation sample

The prognostic influence of the KIR genes found to be significant in the discovery sample was validated using a second independent sample of 174 patients with CML in CP treated with first-line imatinib within the UK multi-center trial (SPIRIT 1). Briefly, the SPIRIT 1 trial is a phase III study in which newly diagnosed CML patients in CP, were randomly allocated to receive either imatinib 400 mg daily or imatinib 800 mg daily. The monitoring schedule employed was similar to the one described above. The median follow-up for the whole group was 34 months and 131 of the patients have been followed for at least 24 months.

KIR genotyping

Genomic DNA was typed for the inhibitory KIR genes KIR2DL1, KIR2DL2, KIR2DL5A (alleles 001 and 005), KIR2DL5B (alleles 002-004, 006 and 007), KIR3DL1, KIR3DL3, the activating KIR genes KIR2DS1, KIR2DS2, KIR2DS3, KIR2DS4, KIR2DS5, KIR3DS1, the framework genes KIR2DL3, KIR2DL4, KIR3DL2, KIR3DP1 and the pseudogene KIR2DP1. Briefly, genotyping was performed using PCR amplification with two locus sequence-specific primers (PCR-SSP) (Invitrogen, Milwaukee, WI, USA). The primer sets amplified alleles described by the International Nomenclature Committee of the World Health Organization (http://www.ebi.ac.uk). Internal control primers that amplify an 800 bp-fragment and a 200 bp-fragment of the alleles of KIR3DP1 (framework gene expressed in nearly all haplotypes) were used to confirm robust PCR amplifications. Amplification was performed in 10 μl of PCR mix containing 500 ng of genomic DNA and 5 U/μl Taq polymerase (Bioline, London, UK). Cycling was performed as follows: 30 cycles of 94° for 20 s, 63 ° for 20 s and 72 ° for 90 s. PCR products were then electrophoresed in 2% agarose gels containing ethidium bromide and the products were visualized under ultraviolet light.

HLA typing

High-resolution HLA typing was performed with the use of PCR with sequence-specific primer and reference strand conformational analysis using in-house primers. HLA-C alleles were classified as group 1 and group 2 based on the amino acid at position 80. Group 1: Cw*01 (02, 03), Cw*03 (02, 03, 041), Cw*07 (01, 02, 03, 04, 05, 06), Cw*08 (01, 02, 03), Cw*12 (021, 022, 03, 06), Cw*14 (002, 03), Cw*16 (01, 03, 041) and group 2: Cw*02 (021, 022, 023, 024) Cw*04 (01), Cw*05 (01), Cw*06 (02), Cw*07 (07), Cw*12 (041, 042, 05), Cw*15 (02, 03, 04, 051, 052), Cw*16 (02), Cw*17 (01, 02), Cw*18 (01, 02) alleles are characterized by the presence of an Asn residue or a Lys, respectively. Group 1 HLA-Cw motifs are the ligands for KIR2DL2 and KIR2DL3, whereas group 2 motifs are ligands for KIR2DL1.

Statistical methods

Probabilities of overall survival (OS) and progression-free survival (PFS) were calculated using the Kaplan–Meier method. PFS was defined as survival without evidence of progression to accelerated or blast phase disease.22 The probability of cytogenetic response was calculated using the cumulative incidence procedure, where cytogenetic response was the event of interest and death and disease progression were the competitors. For cytogenetic response analysis, patients were censored at discontinuation of imatinib, for PFS and OS patients were analyzed in an intention to treat basis. Univariate analyses to identify prognostic factors for survival, PFS and cytogenetic response were carried out using the log-rank test. In the validation study the prognostic influence of the KIR genes was analyzed using a log-rank test stratified by allocated therapy. Variables found to be significant at the P<0.10 level were entered into a proportional hazards regression analysis; a forward stepping procedure was employed to find the best model. We did not analyze the prognostic influence of an individual KIR gene (see below) if the gene prevalence was <10% of the population (15 patients). P-values were two sided.

Results

Discovery study

The discovery study included 166 patients with CML-CP who received imatinib as first-line treatment within 6 months of diagnosis. During follow-up 119 (71.7%) patients achieved CCyR, 10 (6.0%) progressed to advanced phase disease and 7 (4.2%) died. The 2-year cumulative incidence of CCyR was 72.2% and the 2-year probabilities of PFS and OS were 95.7% and 97.3%, respectively.

KIR genotype and CCyR, PFS and OS

The KIR gene frequencies are summarized in Table 2. The 2-year probabilities of achieving CCyR, PFS and OS according to the KIR genotype are shown in Table 2. The presence of the activating KIR genes KIR2DS1, KIR2DS3 and the inhibitory KIRs KIR2DL5A and KIR2DL5B alleles, and absence of the activating KIR gene KIR2DS4 were significantly associated with a lower probability of achieving CCyR (Table 2). The presence of KIR2DL5A, KIR2DS1 KIR2DS5, KIR3DS1 were associated with significantly shorter 2-year probabilities of PFS and OS (Table 2). There was no significant influence of KIR haplotype, or the number of activating KIRs or inhibitory KIRs on the probability of achieving CCyR, PFS and OS (data not shown).

Table 2 Association between individual KIRs and achievement of CCyR, progression-free survival and overall survival at 2 years on univariate analysis

As reported previously, we found a strong linkage disequilibrium between several genes.23, 24, 25 The following genes were in strong positive linkage disequilibrium: KIR2DS1, KIR2DL5A, KIR2DS3, KIR2DS5 and KIR3DS1 (P<0.0001 in all cases). On the other hand KIR2DS1 and KIR2DS4 were in perfect negative linkage disequilibrium (P<0.001).

We also examined the prognostic influence of the variables shown in Table 1 on the 2-year probabilities of achieving CCyR, PFS and OS. Sokal risk score26, 27 at diagnosis was the only predictive variable for achievement of CCyR; the 2-year probability of CCyR for the low, intermediate and high Sokal risk groups were 82.6, 69.8 and 62.1%, respectively, P=0.04. None of the variables (including Sokal risk score) were found to predict for PFS and OS.

We carried out a multivariate analysis for each outcome (CCyR, PFS and OS) including the KIR genes identified in the univariate analysis and for the CCyR analysis we also included the Sokal risk score. The presence of KIR2DS1 (RR=1.51, P=0.03) and Sokal risk score (low risk RR=1, intermediate risk RR=1.53, P=0.04 and high risk RR=1.69, P=0.034) were the only independent predictors for failure to achieve CCyR. Furthermore, the presence of KIR2DS1 alleles within a patient's KIR repertoire was the only independent predictor for shorter PFS (RR=3.1, P=0.03 and OS (RR=2.6, P=0.04). Figure 1 shows the probabilities of achieving CCyR and OS according to KIR2DS1 status.

Figure 1
figure1

The 2-year probabilities of CCyR and OS according to KIR2DS1 genotype on multivariate analysis. The 100 patients who were KIR2DS1 positive (broken line) had a significantly lower probability of achieving CCyR (a) and lower overall survival (b) than the 66 patients (solid line) who were KIR2DS1 negative, namely 65.6 vs 82.3% (P=0.03) and 92.3 vs 100% (P=0.04), respectively.

Validation study

In order to validate our results we explored the influence of the KIR genotype on the achievement of CCyR and on the probabilities of PFS and OS in an independent patient cohort. KIR genotype was determined in 174 patients with CML-CP treated with first-line imatinib in the SPIRIT 1 trial (see methods section). The 65 patients who were KIR2DL5A allele positive had significantly lower 2-year probabilities of achieving CCyR, PFS and OS compared with the 109 patients who were negative for this gene, namely 80.4 vs 86.8% (therapy stratified log-rank test=0.03), 85.7 vs 98.1% (therapy stratified log-rank test=0.007) and 94.3 vs 100% (therapy stratified log-rank test=0.017), respectively. Furthermore, the 66 KIR2DS1-positive patients had lower probabilities of achieving CCyR and lower PFS and OS than the 106 KIR2DS1-negative patients, namely 76.9 vs 87.9% (therapy stratified log-rank test=0.003), 85.3 vs 98.1% (therapy stratified log-rank test=0.007) and 94.4 vs 100% (therapy stratified log-rank test=0.015), respectively (in eight patients KIR2DS1 status could not be accurately determined) (Figure 2). In accord with our finding in the initial patient cohort, on multivariate analysis of the SPIRIT 1 patients, KIR2DS1 positivity remained the only independent predictor for all three outcomes.

Figure 2
figure2

The 2-year probabilities of CCyR and OS according to the KIR2DS1 genotype in the validation sample. The 66 KIR2DS1-positive patients (broken line) had a lower probability of achieving CCyR (a) and lower OS (b) than the 106 KIR2DS1-negative patients (solid line), namely 76.9 vs 87.9% (P=0.003) and 94.4 vs 100% (P=0.015), respectively.

Influence of KIR2DS1 and HLA-C on CCyR

Because a KIR2DL1 interaction with group 2 HLA-C molecules on target cells would theoretically inhibit an activating signal mediated by KIR2DS1, we hypothesized that any effect of KIR2DS1 would be greatest among individuals who are missing group 2 HLA-C ligand for KIR2DL1. Therefore, we determined the various combinatorial frequencies of KIR2DS1 with group 2 HLA-C alleles in the discovery plus the validation patients (n=340). We subdivided patients into three groups on the basis of KIR2DS1/2DL1 genotype and group 2 HLA-C alleles (C2) genotype: group I: KIR2DS1 −ve, KIR2DL1+ve and C2+ve (n=80); group II: KIR2DS1 +ve, KIR2DL1+ve and C2+ve (n=85); and group III: KIR2DS1 +ve, KIR2DL1+ve and C2−ve (n=46). The probability of achieving CCyR at 2 years for patients in group I was 98.6% compared with 74.8% for group II and 64.1% in group III (P=0.00006; Figure 3). These data suggest that the association between the activating gene KIR2DS1 and CCyR is even more significant when the HLA ligand for the closely related inhibitory receptor KIR2DL1 is absent.

Figure 3
figure3

Association between KIR2DS1/KIR2DL1 and HLA-C genotype. Patients who are KIR2DS1+ve/KIR2DL1+ve/C2−ve (open boxes) have significantly lower 2-year probability of CCyR compared with KIR2DS1+ve/KIR2DL1+ve/C2+ve (black circles) and KIR2DS1−ve/KIR2DL1+ve/C2+ve patients (solid line) (P=0.00006).

Discussion

CML is susceptible to immune effector responses and NK cells have well documented anti-leukemia activity.28, 29 Indeed, the association between KIR genotype and outcome following allo-SCT for myeloid leukemia has been shown by a number of groups.30, 31, 32, 33 We recently reported on the relationship between KIR genotype and outcome in the setting of autologous SCT for myeloma.34 Using a large homogeneous cohort of newly diagnosed patients with CML-CP treated with imatinib, we show here that patients carrying the activating KIR gene KIR2DS1 have a significantly lower probability of achieving CCyR, and lower 2-year probabilities of PFS and OS compared with those who are KIR2DS1 negative. On the other hand, absence of KIR2DS1 was associated with better response to imatinib. This effect was independent of Sokal risk score and was validated in a second independent large cohort of patients.

On univariate analysis, five KIR genes were found to have prognostic implications in CML patients on TKI therapy. The presence of KIR2DS1, KIR2DS3, KIR2DL5A and KIR2DL5B and the absence of KIR2DS4 were associated with significantly lower probabilities of achieving CCyR at 2 years. Two of these genes (KIR2DS1 and KIR2DL5A) were also associated with reduced probabilities of PFS and OS. As previously reported, KIR2DS1, KIR2DL5A, KIR2DS3, KIR2DL5A and KIR2DL5B genes were found to be in strong linkage disequilibrium.35, 36 On multivariate analysis, KIR2DS1, a gene with a prevalence of up to 50% in the population, together with the Sokal risk group, were found to be the independent predictors for disease response (failure to achieve CCyR, RR=1.51, P=0.03) Furthermore, this gene was the only predictor for shorter PFS and OS (RR=3.1, P=0.03 and RR=2.6, P=0.04, respectively).

This association of KIR2DS1 with a reduced probability of achieving CCyR at 24 months and inferior PFS and OS in the discovery population, was also observed in the validation cohort, which included newly diagnosed patients with CML treated on the multi-center SPIRIT 1 trial, namely 76.9 vs 87.9% (P=0.003), 85.3 vs 98.1% (P=0.007) and 94.4 vs 100% (P=0.015), respectively. This is especially relevant as the patients in the validation set were randomly assigned to treatment with different doses of imatinib, underscoring the predictive role of KIR2DS1 on outcome of CML, irrespective of the dose of imatinib that the patient may have received.

The impact of the KIR2DS1 genotype on the probability to achieve CCyR was even greater when the expression of this gene was considered together with the presence or absence of the ligand for the corresponding inhibitory KIR receptor (KIR2DL1), suggesting that in the presence of the HLA ligand, the corresponding inhibitory KIR may neutralize the effect of the activating KIR. The same phenomenon has also reported in psoriasis where subjects with activating KIR2DS1 genes were found to be significantly more susceptible to developing psoriatic arthritis when the HLA ligand for the homologous inhibitory receptors, KIR2DL1 was missing.37 Activating KIR molecules are known to bind with reduced affinity to HLA molecules compared with that observed for inhibitory KIRs, perhaps explaining the observed dominance of inhibition over activation of NK cells. Indeed, KIR2DS1 has significantly lower affinity for C2 group molecules than KIR2DL1.37, 38, 39 The increased susceptibility effect of KIR2DS1 in the absence of group 2 HLA-C alleles suggests that ligands for KIR2DS1 may not be the same HLA molecules recognized by homologous KIR2DL1 receptors. Data from mice indicate that viral proteins structurally similar to class I molecules serve as ligands for activating receptors on NK cells,40, 41, 42, 43 raising the possibility that activating receptors in humans may also recognize foreign molecules on target cells as opposed to recognition of self class I molecules. It is therefore possible that KIR2DS1 can recognize non-MHC molecules, such as aberrantly expressed normal cell surface proteins (for example, leukemia-associated antigens), or complexes of CML-derived peptides bound to MHC class I molecules.

The mechanism by which KIR2DS1 affects outcome in patients with CML-CP on imatinib is unclear. NK cells exert cytotoxicity against CML progenitors and kill quiescent CML stem cells.44, 45 However, the interaction between TKI and NK cells is not fully elucidated. Imatinib has been shown to act on the c-kit signaling pathway of host myeloid dendritic cell subsets to promote NK cell IFN-γ secretion and cytotoxicity.46 On the other hand imatinib has been shown to impair NK anti-leukemic activity, possibly by downregulating expression of ligands for NKG2D, an NK receptor important in mediating an anti-tumor response.47 It is possible that imatinib interacts differentially with individual NK subsets or their specific ligands to abrogate or enhance any NK anti-tumor function. Interestingly, NK cells have been shown to undergo clonal expansion in patients receiving TKI.48 Recently NK cells expressing KIR2DS1 were shown to secrete transforming-growth factor-β, which has an inhibitory effect on NK-mediated target lysis and on pro-inflammatory cytokine production.49 Transforming growth factor-β can modulate antigen-presenting cell function and lymphocyte differentiation, leading to expansion of regulatory T-cells and the suppression of auto-reactive T-cells.50 Furthermore, transforming growth factor-β was recently shown to inhibit Akt signaling, a suppressor of the forkhead O transcription factor FOXO3a, in the CML leukemia-initiating cells51 and may represent an important mechanism for the CML leukemia-initiating cells to survive imatinib. It is, however, possible that rather than being directly involved in the disease process, KIR2DS1 may be simply a surrogate marker for another neighboring gene that is directly involved in CML response to TKI.

In conclusion, our data demonstrate that KIR2DS1 may predict response to imatinib and identify patients at greater risk of treatment failure. KIR immunogenetics further supports a role for NK-mediated immunological responses in the control of disease response in CML.

References

  1. 1

    Druker BJ, Guilhot F, O’Brien SG, Gathmann I, Kantarjian H, Deininger MW et al. Five-year follow-up of patients receiving imatinib for chronic myeloid leukemia. N Engl J Med 2006; 355: 2408–2417.

  2. 2

    Hughes TP, Kaeda J, Branford S, Rudzki Z, Hochhaus A, Hensley ML et al. Frequency of major molecular responses to imatinib or interferon alfa plus cytarabine in newly diagnosed chronic myeloid leukemia. N Engl J Med 2003; 349: 1423–1432.

  3. 3

    Kolb HJ, Schmid C, Barrett AJ, Schendel DJ . Graft-versus-leukemia reactions in allogeneic chimeras. Blood 2004; 103: 767–776.

  4. 4

    Caligiuri MA . Human natural killer cells. Blood 2008; 112: 461–469.

  5. 5

    Lanier LL . NK cell recognition. Annu Rev Immunol 2005; 23: 225–274.

  6. 6

    Vivier E, Nunes JA, Vely F . Natural killer cell signaling pathways. Science 2004; 306: 1517–1519.

  7. 7

    Clausen J, Wolf D, Petzer AL, Gunsilius E, Schumacher P, Kircher B et al. Impact of natural killer cell dose and donor killer-cell immunoglobulin-like receptor (KIR) genotype on outcome following human leucocyte antigen-identical haematopoietic stem cell transplantation. Clin Exp Immunol 2007; 148: 520–528.

  8. 8

    Cooley S, Trachtenberg E, Bergemann TL, Saeteurn K, Klein J, Chap T 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.

  9. 9

    Kroger N, Binder T, Zabelina T, Wolschke C, Schieder H, Renges H et al. Low number of donor activating killer immunoglobulin-like receptors (KIR) genes but not KIR-ligand mismatch prevents relapse and improves disease-free survival in leukemia patients after in vivo T-cell depleted unrelated stem cell transplantation. Transplantation 2006; 82: 1024–1030.

  10. 10

    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.

  11. 11

    Sconocchia G, Lau M, Provenzano M, Rezvani K, Wongsena W, Fujiwara H et al. The antileukemia effect of HLA-matched NK and NK-T cells in chronic myelogenous leukemia involves NKG2D-target-cell interactions. Blood 2005; 106: 3666–3672.

  12. 12

    Yong AS, Keyvanfar K, Hensel N, Eniafe R, Savani BN, Berg M et al. Primitive quiescent CD34+ cells in chronic myeloid leukemia are targeted by in vitro expanded natural killer cells, which are functionally enhanced by bortezomib. Blood 2009; 113: 875–882.

  13. 13

    Kreutzman A, Juvonen V, Kairisto V, Ekblom M, Stenke L, Seggewiss R et al. Mono/oligoclonal T and NK cells are common in chronic myeloid leukemia patients at diagnosis and expand during dasatinib therapy. Blood 2010; 116: 772–782.

  14. 14

    Borg C, Terme M, Taieb J, Menard C, Flament C, Robert C et al. Novel mode of action of c-kit tyrosine kinase inhibitors leading to NK cell-dependent antitumor effects. J Clin Invest 2004; 114: 379–388.

  15. 15

    Cebo C, Da RS, Wittnebel S, Turhan AG, Abdelali J, Caillat-Zucman S et al. The decreased susceptibility of Bcr/Abl targets to NK cell-mediated lysis in response to imatinib mesylate involves modulation of NKG2D ligands, GM1 expression, and synapse formation. J Immunol 2006; 176: 864–872.

  16. 16

    Kantarjian HM, Talpaz M . Definition of the accelerated phase of chronic myelogenous leukemia. J Clin Oncol 1988; 6: 180–182.

  17. 17

    O’Brien SG, Deininger MW . Imatinib in patients with newly diagnosed chronic-phase chronic myeloid leukemia. Semin Hematol 2003; 40: 26–30.

  18. 18

    Druker BJ, Guilhot F, O’Brien SG, Gathmann I, Kantarjian H, Deininger MW et al. Five-year follow-up of patients receiving imatinib for chronic myeloid leukemia. N Engl J Med 2006; 355: 2408–2417.

  19. 19

    O’Brien SG, Deininger MW . Imatinib in patients with newly diagnosed chronic-phase chronic myeloid leukemia. Semin Hematol 2003; 40: 26–30.

  20. 20

    National Institutes of Health National Cancer Institute. Cancer Therapy Evaluation Program: Common Toxicity Criteria. National Institutes of Health National Cancer Institute: Bethesda, MD, 1998.

  21. 21

    Baccarani M, Castagnetti F, Gugliotta G, Palandri F, Soverini S . Response definitions and European Leukemianet Management recommendations. Best Pract Res Clin Haematol 2009; 22: 331–341.

  22. 22

    Kantarjian HM, Talpaz M . Definition of the accelerated phase of chronic myelogenous leukemia. J Clin Oncol 1988; 6: 180–182.

  23. 23

    Velickovic M, Velickovic Z, Dunckley H . Diversity of killer cell immunoglobulin-like receptor genes in Pacific Islands populations. Immunogenetics 2006; 58: 523–532.

  24. 24

    Hsu KC, Chida S, Geraghty DE, Dupont B . The killer cell immunoglobulin-like receptor (KIR) genomic region: gene-order, haplotypes and allelic polymorphism. Immunol Rev 2002; 190: 40–52.

  25. 25

    Single RM, Martin MP, Meyer D, Gao X, Carrington M . Methods for assessing gene content diversity of KIR with examples from a global set of populations. Immunogenetics 2008; 60: 711–725.

  26. 26

    Sokal JE, Cox EB, Baccarani M, Tura S, Gomez GA, Robertson JE et al. Prognostic discrimination in “good-risk” chronic granulocytic leukemia. Blood 1984; 63: 789–799.

  27. 27

    de Lavallade H, Apperley JF, Khorashad JS, Milojkovic D, Reid A, Bua M et al. Imatinib for newly diagnosed patients with chronic myeloid leukemia: incidence of sustained responses in an intention-to-treat analysis. J Clin Oncol 2008; 26: 3358–3363.

  28. 28

    Grzywacz B, Miller JS, Verneris MR . Use of natural killer cells as immunotherapy for leukaemia. Best. Pract Res Clin Haematol 2008; 21: 467–483.

  29. 29

    Sconocchia G, Lau M, Provenzano M, Rezvani K, Wingsena W, Fujiwara H et al. The antileukemia effect of HLA-matched NK and NK-T cells in chronic myelogenous leukemia involves NKG2D-target-cell interactions. Blood 2005; 106: 3666–3672.

  30. 30

    Clausen J, Wolf D, Petzer AL, Gunsilius E, Schumacher P, Kircher B et al. Impact of natural killer cell dose and donor killer-cell immunoglobulin-like receptor (KIR) genotype on outcome following human leucocyte antigen-identical haematopoietic stem cell transplantation. Clin Exp Immunol 2007; 148: 520–528.

  31. 31

    Giebel S, Locatelli F, Lamparelli T, Velardi, 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.

  32. 32

    Miller JS, Cooley S, Parham P, Farag SS, Verneris MR, Mc Queen KL 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: 5058–5061.

  33. 33

    Ruggeri L, Capanni M, Urbani E, Perruccio, Shlomick WD, Tosti A et al. Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic transplants. Science 2002; 295: 2097–2100.

  34. 34

    Gabriel IH, Sergeant R, Szydlo R, Apperley JF, DeLavallade H, Alsuliman A et al. Interaction between KIR3DS1 and HLA-Bw4 predicts for progression-free survival following autologous stem cell transplantation in patients with multiple myeloma. Blood 2010; 116: 2033–2039.

  35. 35

    Hsu KC, Chida S, Geraghty DE, Dupont B . The killer cell immunoglobulin-like receptor (KIR) genomic region: gene-order, haplotypes and allelic polymorphism. Immunol Rev 2002; 190: 40–52.

  36. 36

    Single RM, Martin MP, Meyer D, Gao X, Carrington M . Methods for assessing gene content diversity of KIR with examples from a global set of populations. Immunogenetics 2008; 60: 711–725.

  37. 37

    Martin MP, Nelson G, Lee JH, Pellett F, Gao X, Wade J et al. Cutting edge: susceptibility to psoriatic arthritis: influence of activating killer Ig-like receptor genes in the absence of specific HLA-C alleles. J Immunol 2002; 169: 2818–2822.

  38. 38

    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 USA 2005; 102: 13224–13229.

  39. 39

    Biassoni R, Pessino A, Malaspina A, Cantoni, Bottino C, Sivori S et al. Role of amino acid position 70 in the binding affinity of p50.1 and p58.1 receptors for HLA-Cw4 molecules. Eur J Immunol 1997; 27: 3095–3099.

  40. 40

    Arase H, Mocarski ES, Campbell AE, Hill AB, Lanier LL . Direct recognition of cytomegalovirus by activating and inhibitory NK cell receptors. Science 2002; 296: 1323–1326.

  41. 41

    Daniels KA, Devora G, Lai WC, O’Donnell CL, Bennett M, Welsh RM . Murine cytomegalovirus is regulated by a discrete subset of natural killer cells reactive with monoclonal antibody to Ly49 h. J Exp Med 2001; 194: 29–44.

  42. 42

    Smith HRC, Heusel JW, Mehta IK, Kim S, Dorner BG, Naidenko OV et al. Recognition of a virus-encoded ligand by a natural killer cell activation receptor. Proc Natl Acad Sci USA 2002; 99: 8826–8831.

  43. 43

    Vivier E, Biron CA . A pathogen receptor on natural killer cells. Science 2002; 296: 1248–1249.

  44. 44

    Sconocchia G, Lau M, Provenzano M, Rezvani K, Wonsenga W, Fujiwara H et al. The antileukemia effect of HLA-matched NK and NK-T cells in chronic myelogenous leukemia involves NKG2D-target-cell interactions. Blood 2005; 106: 3666–3672.

  45. 45

    Yong AS, Keyvanfar K, Hensel N, Enaife R, Savani BN, Berg M et al. Primitive quiescent CD34+ cells in chronic myeloid leukemia are targeted by in vitro expanded natural killer cells, which are functionally enhanced by bortezomib. Blood 2009; 113: 875–882.

  46. 46

    Borg C, Terme M, Taieb J, Menard C, Flament C, Robert C et al. Novel mode of action of c-kit tyrosine kinase inhibitors leading to NK cell-dependent antitumor effects. J Clin Invest 2004; 114: 379–388.

  47. 47

    Salih J, Hilpert J, Placke T, Grunebach F, Steinle A, Salih HR et al. The BCR/ABL-inhibitors imatinib, nilotinib and dasatinib differentially affect NK cell reactivity. Int J Cancer 2010; 127: 2119–2128.

  48. 48

    Mustjoki S, Ekblom M, Arstila TP, Dybedal L, Epling-Burnette PK, Guihot F et al. Clonal expansion of T/NK-cells during tyrosine kinase inhibitor dasatinib therapy. Leukemia 2009; 23: 1398–1405.

  49. 49

    Ghio M, Contini P, Negrini S Mazzei C, Zocchi MR, Poggi A . Soluble HLA-I-mediated secretion of TGF-beta1 by human NK cells and consequent down-regulation of anti-tumor cytolytic activity. Eur J Immunol 2009; 39: 3459–3468.

  50. 50

    Nakamura K, Kitani A, Strober W . Cell contact-dependent immunosuppression by CD4(+)CD25(+) regulatory T cells is mediated by cell surface-bound transforming growth factor beta. J Exp Med 2001; 194: 629–644.

  51. 51

    Naka K, Hoshii T, Muraguchi T, Tadokoro, Ooshio T, Kondo Y et al. TGF-beta-FOXO signalling maintains leukaemia-initiating cells in chronic myeloid leukaemia. Nature 2010; 463: 676–680.

Download references

Acknowledgements

We would like to acknowledge the participating centers in the SPIRIT 1 trial, the SPIRIT study Team especially Caroline Hodgson, Claire Oyston, Lynn Seeley, Wendy Banks, Meg Buckley and the support of the NCRN CML working group, United Kingdom. We acknowledge the support of the National Institute for Health Research (NIHR) Biomedical Research Center (BRC), United Kingdom. This work was supported by the NIHR BRC London, (grant no. P31514) and Leuka registered charity, 286231.

Author information

Correspondence to K Rezvani.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • chronic myeloid leukemia
  • natural killer cells
  • killer immunoglobulin-like receptors
  • imatinib response

Further reading