Human Leukocyte Antigen and Killer Immunoglobulin-like...

1

Human Leukocyte Antigen and Killer Immunoglobulin-like receptor genes as outcome

predictors of hepatitis C virus-related hepatocellular carcinoma

Elisabetta Cariani1, Massimo Pilli2, Alessandro Zerbini3, Cristina Rota1, Andrea Olivani2, Paola

Zanelli4, Adele Zanetti4, Tommaso Trenti1, Carlo Ferrari2, Gabriele Missale2

1Clinical Pathology-Toxicology, Ospedale S. Agostino-Estense, Modena

2U.O. Infectious Diseases and Hepatology, 4U.O. Medical Genetics, Azienda Ospedaliero-

Universitaria di Parma

3Clinical Immunology, Allergy and Advanced Biotechnologies Unit, Department of Laboratory

Medicine, Azienda Ospedaliera ASMN, Istituto di Ricovero e Cura a Carattere Scientifico, Reggio

Emilia, Italy

Running title: KIR and HLA loci in HCV-related HCC

Keywords:

Natural Killer cells; Surgical resection; Radiofrequency thermal ablation; CD56; CD107a

Grant Support

Supported by a grant from Fondazione CARIPARMA (Parma, Italy); by the grant RBAP10TPXK

from Fondo per gli Investimenti della Ricerca di Base (Ministry of Education, University and

Research, Italy).

Corresponding Author

Gabriele Missale

U.O. Malattie Infettive ed Epatologia

Research. on February 14, 2019. © 2013 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on August 12, 2013; DOI: 10.1158/1078-0432.CCR-13-0986

http://clincancerres.aacrjournals.org/

2

Azienda Ospedaliero-Universitaria di Parma

Tel +39 0521 702310 – Fax +39 0521 703857

e-mail: [email protected]

Disclosure of Potential Conflict of Interest

No potential conflicts of interest were disclosed

Electronic word count: 4699

Number of figures : 3

Number of tables: 3

Authors’ Contributions

Conception and design: E. Cariani, C. Ferrari, G. Missale

Development of methodology: E. Cariani, M. Pilli, G. Missale

Acquisition of data: M. Pilli, C. Rota, A. Olivani, P. Zanelli, A. Zanetti

Analysis and interpretation of data: E. Cariani, A. Zerbini, C. Ferrari, G. Missale

Writing, review, and/or revision of manuscript: E. Cariani, M. Pilli, A. Zerbini, C. Rota, A.

Olivani, P. Zanelli, A. Zanetti, T. Trenti, C. Ferrari, G. Missale

Administrative, technical or material support: M. Pilli, C. Rota

Study supervision: T. Trenti, C. Ferrari, G. Missale




3

Translational relevance

Natural killer (NK) cells play a major role in anti-tumor immune response. The genotypes of NK

Killer Immunoglobulin-like receptors (KIRs) and of their HLA ligands are related to the

development of hepatocellular carcinoma (HCC) in patients with HCV infection. A better

understanding of the role played by the immunogenotypic background over HCC outcome might

represent a major breakthrough to improve the clinical management of HCC patients.

Our work shows that the immunogenetic profile is associated with NK-cell function and represents

a powerful outcome predictor in HCV-linked HCC. These results support a central role of NK cells

in the immune response against HCC with important implications in terms of development of

immunotherapeutic approaches and individualized monitoring schemes after treatment of early

HCC.




4

Abstract

Purpose: We evaluated the impact of the Killer Immunoglobulin-like Receptors (KIRs) of natural

killer (NK) cells and of their Human Leukocyte Antigen (HLA) ligands over the clinical outcome of

HCV-related hepatocellular carcinoma (HCC) after curative treatment by either surgical resection

(SR) or radiofrequency thermal ablation (RTA).

Experimental Design: Sixty-one consecutive patients with HCV-related HCC underwent KIR

genotyping and HLA typing. A phenotypic/functional characterization of NK-cells was carried out

in patients with different KIR/KIR-ligand genotype.

Results: Activating KIR2DS5 was associated with significantly longer time to recurrence (TTR)

and overall survival (OS) (p<0.03 each). Homozygous HLA-C1 (p<0.02) and HLA-Bw4I80

(p<0.05) were expressed by patients with significantly better OS, while HLA-C2 (p<0.02) and

HLA-Bw4T80 (p<0.01) were associated with a worse OS. Multivariate analysis identified as

parameters independently related to TTR the type of treatment (SR vs RTA) (p<0.03) and HLA-C1

(p<0.03), whereas only KIR2DS5 was an independent predictor of longer OS (p<0.05). Compound

KIR2DL2-C1 and KIR3DS1-Bw4T80 genotypes were associated with better TTR (p<0.03) and

worse OS (p=0.02), respectively. A prevalent cytotoxic (CD56dim) NK phenotype was detected in

patients with both longer TTR and OS. Cytotoxic capacity measured by upregulation of CD107a

was significantly higher in subjects with HLA-C1 alone or combined with KIR2DL2/KIR2DL3.

Conclusions: These results support a central role of NK-cells in the immune response against HCC,

providing a strong rationale for therapeutic strategies enhancing NK response and for individualized

post-treatment monitoring schemes.




5

Introduction

Natural killer (NK) cells recognize target cells through interaction of surface inhibitory and

activating receptors with their ligands. NK-cells have been divided into two major populations, one

mainly cytotoxic in function (CD56dimCD16+) and the other mainly involved in cytokine secretion

(CD56bright, CD16dim or negative) (1).

NK cell receptors belong to two main families: the C-type lectins-like (NKG2) receptors and the

immunoglobulin (Ig)-like superfamily, including the killer cell Ig-like receptors (KIRs). KIR

genotypes can be grouped into haplotypes A and B, mainly including inhibitory (A) or activating

(B) KIRs, respectively (2). The B haplotypes contain variable numbers and combinations of KIR

genes.

The fine tuning of KIR-ligand interactions is the result of multi-level regulatory mechanisms

relying both on the quantity of KIR/HLA molecules expressed on NK and target cells, and on the

affinity of KIR-HLA interactions. Inhibitory KIR2DL1 receptor recognizes alleles of HLA-C with

lysine at position 80 (HLA-C2), whereas KIR2DL2 and KIR2DL3, segregating as alleles of a single

locus, specifically bind HLA-C alleles with asparagine at position 80 (HLA-C1) (3,4) with different

binding affinity (5): KIR2DL3/HLA-C1 interaction is thought to be weaker than both

KIR2DL2/HLA-C1 and KIR2DL1/HLA-C2. In addition KIR2DL2, and to a lesser extent

KIR2DL3, bind with low affinity HLA-C2 (5). KIR3DL1 recognizes the Bw4 motif of HLA-B

alleles (6), generating a stronger inhibitory signal when an isoleucine residue is present at position

80 (Bw4I80). Due to the relevant homology of activating KIR2DS2 and KIR3DS1 to inhibitory

KIR2DL2/DL3 and KIR3DL1, respectively, it has been speculated that they might share the same

ligands as the inhibitory KIR counterparts, but this assumption has not been demonstrated yet. By

contrast, KIR2DS1 has been shown to bind HLA-C2 as KIR2DL1, although with lower affinity (7).

NK-cells recognize and eliminate cells that fail to express self HLA molecules, such as virus-

infected and transformed cells. In hepatitis C virus (HCV) infection, impaired NK cell frequency

and function has been reported (8) and specific KIR/ligand genotypes have been implicated in the




6

clinical evolution and therapeutic response: the KIR2DL3/HLA-C1 genotype has been associated

with the resolution of infection (9-11) as KIR3DS1 and HLA-Bw4, although with a weak protective

effect (9). By contrast, homozygous HLA-C2 (9) and KIR2DS3, in the presence of HLA-C2 (12)

were more frequent in patients with chronic hepatitis C compared to individuals with spontaneously

resolved infection. In patients with acute HCV infection, increased NK-cells cytotoxicity was

present in subjects expressing HLA-C1-specific KIR2DL2/3 and in particular in self-limited

infection (13). Other studies reported over-representation of homozygous KIR2DL3/HLA-C1 in

sustained responders (11,14) and of HLA-C2C2 in patients resistant to treatment (15).

Functional impairment of NK-cells has been observed in patients with hepatocellular carcinoma

(HCC) (16). The KIR3DS1/HLA-Bw4I80 genotype and HLA-C1 were interpreted as protective

against the development of HCV-related HCC (17), but no data are available about the potential

relationship between KIR/HLA genotypes and the prognosis of HCC after treatment. We have

evaluated the impact of the immunogenetic host background over the clinical outcome of HCV-

related HCC after curative treatment by either surgical resection (SR) or radiofrequency thermal

ablation (RTA).




7

Materials and Methods

Patients

We evaluated 61 consecutive Caucasian HCC patients with HCV-related liver disease, that

underwent curative treatment by either SR or RTA at the University Hospital of Parma, Italy. HCC

diagnosis was made by ultrasonography (US) and computed tomography (CT) or magnetic

resonance imaging (MRI) in selected cases. The type of treatment was decided on the basis of liver

function (Child-Pugh score), comorbidities, age and location of HCC nodules. Patients with early

HCC were evaluated by a multidisciplinary group (interventional radiologist, hepatologist and

surgeon) in order to decide treatment allocation based on liver function, portal hypertension,

number, site and size of HCC lesions. Patients with one or two lesions less than 15 mm in diameter

were treated with percutaneus alcohol injection. All patients had not been previously treated for

HCC.

All patients undergoing liver resection were within Child-A score. HBsAg and anti-human

immunodeficiency virus were negative for all cases. All patients had the same postoperative follow-

up based on bidimensional and contrast-enhanced (CE) US every 4 months and dynamic CT or CE

MRI in any case of appearance of new liver nodules more than 1 cm in diameter or arterial

enhancement at the site of previous ablation. The clinico-pathological features of the patients are

shown in Table 1 .

The study was approved by the local ethical committee (Comitato Etico Indipendente of the

Azienda Ospedaliero-Universitaria of Parma, Italy). Patients gave written informed consent to

participate in the study.

KIR genotyping and HLA typing

DNA was extracted from frozen peripheral blood mononuclear cells (PBMC) derived from all

patients using the QIAamp DNA blood Mini Kit (Qiagen, Hilden, Germany). KIR genotypes were

determined by duplex real-time polymerase chain reaction (PCR) (18) on 5 ng DNA in 10 μl




8

containing 1X SsoFastTM EvaGreen® Supermix (Bio-Rad Laboratories, Hercules, CA), 0.3 μM of

each KIR-specific primer, 0.1 μM of each internal control primer, in a Rotor Gene Q thermal cycler

(Qiagen) as follows: 2’ 98°C, 5” 98°C and 10” 62°C for 45 cycles. The PCR products were then

discriminated by melting (2’ 98°C, 10” 75°C, and ramping by 0.5°C/s up to 98°C). The KIR genes

analyzed are shown in Supplementary Figure S1.

HLA typing was performed in all patients by PCR Sequence Specific Priming on genomic DNA. In

all cases showing ambiguities for definition of the HLA-B or C supertypes C1, C2, Bw4, Bw6 and

Bw4 I80/T80 variants, high resolution typing was performed by PCR Sequence Specific

Oligonucleotide Probes (Proimmune Co, Oxford, UK).

Immunostaining of NK-cells

In 38 of 61 patients an aliquot of frozen PBMC obtained the same day or the day before treatment

was available for phenotypic analysis. PBMC were resuspended in RPMI 1640, and 8% human

serum. PBMC (3x105) were stained with CD3-PerCP (BD Biosciences-Pharmingen, San Jose, CA)

and CD56-APC (Miltenyi Biotec, Bergisch Gladbach, Germany). Cells were analyzed on FACSC-

CantoII flow cytometer (BD Biosciences, Franklin Lakes, NJ) by the FACSDiva Software.

Frequency of CD56dim and CD56bright NK-cells was defined for each patient evaluating different

fluorescence intensity of CD3-CD56+ cells.

Evaluation of cytotoxic function and interferon (IFN)-γ production in NK-cells

In 31 of 61 patients an aliquot of frozen PBMC obtained the same day or the day before treatment

was available for functional analysis. For the evaluation of the cytotoxic function, PBMC (1x106)

were incubated 16 hours at 37°C in the presence or absence of 1 ng/ml rhIL-15 (R&D Systems,

Abingdon, UK). PE/Cy5-5-conjugated CD107a (BD Biosciences) was then added with or without

K562 target cells at an E:T ratio of 5:1. After 1 h, 10 µg/ml of brefeldin A (Sigma-Aldrich, St.

Louis, MO) was added and cell were incubated further 3 h at 37°C, then harvested and stained for




9

NK cell surface markers. Percentage of degranulating NK-cells was calculated by subtracting

CD107a staining after incubation without K562 to CD107a staining after incubation with K562

targets.

PBMC (1x106) were incubated for 14 h at 37°C with or without 10 ng/ml rhIL-12 and rhIL-18

(Sigma-Aldrich). Brefeldin A (Sigma-Aldrich) (10 µg/ml) was added for the last 3 h of incubation.

After staining with CD56-APC (Miltenyi Biotec) and CD3-APC/Cy7 (Biolegend, San Diego, CA),

cells were fixed and permeabilised using Fix and Perm medium A and B (Caltag Laboratories,

Burlingame, CA) according to the supplier’s instructions. Cells were stained with anti-IFN-γ

PerCP/Cy5-5 mAb (Biolegend). The proportion of IFN-γ producing cells was determined by

subtracting the percentage of IFN-γ+ cells in unstimulated samples from the one of IL-12/IL-18-

stimulated samples.

Statistical analysis

The differences between groups of continuous variables were analyzed by t-test for unpaired data.

Categorical variables were compared by the Chi-square test or Fisher exact test, as appropriate.

Survival curves were estimated by the Kaplan-Meier method and compared by log-rank test. For

continuous variables, ROC curve analysis was used to define threshold levels discriminating

subgroups of patients with survival/time to recurrence higher or lower than median values, for

subsequent analysis by the log-rank test. Cox proportional hazards regression model was performed

for multivariate survival analysis. Only variables available for all patients and presenting p<0.1 in

the univariate analysis (Table 2) were included in the multivariate model. A p value <0.05 (two-

tailed) was considered significant.




10

Results

Clinical outcome

The study was performed on 61 consecutive Caucasian patients with diagnosis of HCV-related

HCC allocated to treatment by either SR or RTA. (Table 1). The median overall survival (OS) was

43 months and the median time to the first HCC recurrence (TTR) was 17 months. By dividing

patients according to the treatment received, age was significantly higher and liver function worse

in RTA-treated patients, while HCC nodules were on average larger in patients treated by SR.

Subjects that performed SR showed a significantly better OS and TTR compared to patients treated

by RTA (Tables 1 and 2). Female sex was associated with significantly shorter OS, and Child score

with worse TTR; age, number and size of HCC nodules did not affect survival (Table 2). From

these results, survival differences related to treatment seem to be most likely due to the treatment

allocation criteria, since RTA-treated patients were significantly older and had a worse Child score.

Moreover, 3 patients who underwent surgical resection during the same period, died within 2

months from surgery. These patients were not included in the study because the effect of HLA and

KIR associations could not be evaluated.

Individual KIRs and HLA expression and their association with disease progression and overall

survival

KIRs frequencies identified in our patients cohort were comparable to the ones previously reported

in Caucasian patients from Northern Italy (19,20), without any significant difference in the

prevalence of each reported KIR (Supplementary Figure S1).

The 15 patients carrying the KIR2DS5 gene, that is present in a fraction of the KIR B haplotypes,

showed significantly longer TTR and OS (Figure 1). KIR2DS5 is an activating receptor in linkage

disequilibrium with KIR2DS1, that also showed a trend towards a protective effect against

recurrence (Table 2). No other analyzed KIRs were significantly correlated to clinical outcome.

Homozygous HLA-C1 and HLA-Bw4, I80 variant, were associated with significantly better OS,




11

while HLA-C2 and HLA-Bw4T80 were associated with a worse OS (Figure 1 and Table 2).

Multivariate analysis identified as independent parameters associated with outcome the type of

treatment (SR vs RTA) and HLA-C1 for TTR, and only KIR2DS5 for OS (Table 3). By analyzing

the effect of HLA-C1 copy number on survival, we detected that HLA-C1 homozygous patients

showed significantly longer OS compared to HLA-C1 heterozygous patients, and both longer OS

and TTR compared to those homozygous for HLA-C2 (Figure 2A).

Effect of the compound KIR-HLA genotype on disease progression and overall survival

The inhibitory receptors KIR2DL2 and KIR2DL3 share the same ligand HLA-C1. Activating

KIR2DS2 is in strong linkage disequilibrium and highly homologous to KIR2DL2, however

experimental evidence of that KIR2DS2 binds HLA-C1 is not available at present. KIR2DL1 and

KIR2DS1 bind HLA-C2, while KIR3DL1 binds HLA-Bw4, with higher and lower affinity for I80

and T80 variants, respectively (21). Since KIR3DS1 and KIR3DL1 segregate as alleles of the same

locus and show 97% similarity in their extracellular domains, they may share the same HLA ligand.

For the remaining KIRs genotyped in this study, the binding HLA molecule is not known.

Since almost all our patients were positive for both KIR2DL1 and KIR3DL1 (Supplementary

Figure S1), the effect of compound KIR-HLA genotype on disease progression was not evaluated

for these molecules,. Altogether, combined survival analysis was performed for KIR2DL2/2DL3-

C1, KIR2DS1-C2, KIR2DS2-C1 and KIR3DS1-Bw4 (I80 and T80). Significant associations were

found for KIR2DL2-C1 and KIR2DS2-C1 with a better TTR and for KIR3DS1-Bw4T80 with a

worse OS (Figure 2B and Table 2).

NK-cell phenotype and function

First we evaluated whether effector NK-cell phenotype (CD56dim) could be associated with a better

clinical outcome. Cut-off values of CD56dim% able to discriminate patients with TTR or OS higher

than median values (17 and 43 months, respectively) were identified by ROC curve analysis. Indeed




12

patients with a frequency of NK cells with a CD56dim phenotype higher than the above-mentioned

cut-offs showed significantly longer TTR and OS (Figure 3 upper left panels). We then compared

NK-cell function of patients with specific HLA, KIR and HLA-KIR genotypes showing a

significant effect on survival analysis. The distribution of HLA and KIR genotypes in the subgroup

of patients with samples available for functional analysis was comparable to the one of the whole

cohort (data not shown). Cytotoxic capacity, measured by upregulation of CD107a, was

significantly higher in subjects with HLA-C1 alone or associated with KIR2DL2/KIR2DL3 (Figure

3 upper right panel and lower panels). No significant differences could be found for IFN-γ

production (data not shown).




13

Discussion

KIR receptors and their HLA ligands have been implicated as risk factors for several human tumors,

chronic inflammatory diseases, and viral infections (22). A recent genome-wide association study

(23) identified an association between risk of HCC development in patients with chronic hepatitis

and a single nucleotide polymorphism in the gene encoding for MHC-associated chain A (MICA),

the ligand of NKG2D, an activating NK receptor (24). This observation strongly supports a relevant

role for NK-cells in HCC pathogenesis.

In the present study we evaluated the impact of the immunogenetic profile on the clinical outcome

of HCC patients undergoing curative treatment by either SR or RTA. Although both these treatment

strategies are able to eradicate neoplastic nodule(s), the relapse rate is high since, in addition to the

intrahepatic dissemination of primary tumor, de novo HCC may develop in the pre-tumorous

environment of liver cirrhosis. The direct clinical consequences of the underlying liver disease have

also a major impact on OS after treatment. Since the biological characteristics of the tumor and the

host immune response are both involved in HCC progression (25-27), it is likely that the

immunogenetic background determining the NK-cell response may play a role on the risk of relapse

and ultimately on the survival of HCC patients.

A significant association was observed between KIR2DS5, an activating receptor part of haplotype

B, and longer OS and TTR, but the implications of our results could not be further investigated since

the ligand of KIR2DS5 has not been identified yet. Whether KIR2DS5 is a marker for a haplotype

involved in HCC outcome, or whether NK-cells bearing KIR2DS5 play a direct role in the clinical

outcome due to specific functional characteristics remains to be determined. A survival advantage of

patients homozygous for HLA-C1 (HLA-C1C1) and therefore lacking HLA-C2 was also shown,

together with an association of HLA-Bw4 variants I80 and T80 with longer or shorter OS,

respectively. Homozygous HLA-C1, as well as HLA-Bw4I80, has already been reported as

protective against development of HCV-related HCC (17). Our results further support the role of

these genotypes in human HCC, showing their association with survival after curative treatment.




14

Multivariate survival analysis identified as independent parameters associated with outcome the

type of treatment and HLA-C1 for TTR, and only KIR2DS5 for OS. These observations suggest

that the therapeutic approach is a major determinant for disease recurrence, together with the HLA-

C genotype, but does not represent an independent risk factor for OS, where the immunogenetic

profile appears to be the only independent predictive variable.

Previous studies mainly analyzed the implications of the co-expression of inhibitory KIRs and their

HLA ligands, that leads to the maturation of fully competent (“licensed”) NK-cells both in mice and

in humans (28,29). The opposite situation may occur when activating KIRs are co-expressed with

their ligands, as it was shown for KIR2DS1 that recognizes HLA-C2 (30). Indeed the compound

KIR2DS1/HLA-C2 immunogenetic profile would lead to reduced functional competence of NK-

cells. In HCC patients we observed that the concurrent presence of KIR2DL2 and HLA-C1 was

associated with longer TTR, possibly due to the protective effect of NK-cells that had reached full

functional competence upon maturation. Consistent with this view, HLA-Bw4I80 (a high-affinity

ligand of inhibitory KIR3DL1, expressed by 60/61 patients) was associated with longer survival,

whereas the opposite was observed in patients with HLA-Bw4T80, a low affinity ligand of KIR3DL1

(21,31). In addition, the combined presence of KIR3DS1 and HLA-Bw4T80 was associated with

reduced OS. Previous reports suggested a protective effect of HLA-Bw4I80 in association with

activating KIR3DS1 towards the development of HCV-linked HCC (17), the progression of acquired

immunodeficiency syndrome (32) and, with weak effect, the development of chronic HCV infection

(9). Although the binding of KIR3DS1 to HLA-Bw4 has not been convincingly demonstrated so far,

these results suggest that a less effective interaction of KIR3DS1 and HLA-Bw4T80 in HCC patients

might result in reduced NK function and impaired survival. The significance of the association

between KIR2DS2/HLA-C1 compound genotype and longer TTR is uncertain: to date, no

experimental evidence of HLA-C1 binding by KIR2DS2 is available. Therefore, it cannot be ruled

out that this association only derives from the strict linkage disequilibrium between KIR2DL2 and

KIR2DS2.




15

Activating KIR–HLA genotypes appear to increase the risk of developing some virus-related tumors,

such as cervical cancer due to papilloma virus infection and nasopharyngeal carcinoma associated

with Epstein Barr virus infection (33,34). An increased risk of developing HCC was also reported in

patients with chronic HBV infection carrying HLA-C1C1, HLA-Bw4I80 and 22 bp-deleted form of

KIR2DS4 (KIR2DS4/1D) (35), thus suggesting that increased NK cell function might represent a risk

factor for HBV-related HCC. The apparent discrepancy between these observations and those

reported in HCV-linked HCC (17) might derive from differences in the specific immunopathogenic

mechanisms. Since NK-cells exert both antiviral and anti-tumor functions, an enhanced effector

phenotype might be protective against tumor development altogether increasing the pro-

inflammatory anti-viral environment and the consequent tissue damage, both favouring a pre-

neoplastic condition. However, in our study the tumor had already developed and

immunosurveillance by NK-cells would represent an immunological mechanism affecting clinical

outcome.

The impact of the immunogenetic background was further analyzed by phenotypic and functional

characterization of NK-cells. Higher percentage of CD56dim NK cells, indicative of prevalent effector

cytotoxic phenotype, was associated with both longer TTR and OS, and NK-cells from patients

carrying HLA-C1 had significantly higher cytotoxic activity. In addition, the combined presence of

KIR2DL2/KIR2DL3 and HLA-C1 was associated with increased cytotoxic function, further

supporting the relevance of NK-cell licensing for the achievement of full functional competence. The

same association was reported in patients with spontaneously resolving acute HCV infection (13).

This is to our knowledge the first study addressing the relevance of KIR genotype on the prognosis of

HCV-related HCC after curative treatment. Results suggest a central role of NK-cells in the immune

response against HCC, thus supporting the rationale for developing immunomodulatory strategies

aimed at the enhancement of NK response (36) especially in patients showing a favorable

immunogenotypic background. In addition, NK cell phenotype and function may provide information




16

useful to predict outcome and tailor follow-up strategies after HCC treatment. Thus our results, if

confirmed on larger series of patients, will be relevant for the implementation of individualized post-

treatment monitoring schemes and novel therapeutic strategies in patients with HCC.




17

Figure Legends

Figure 1.

Individual KIRs and HLA molecules associated with better TTR and/or OS.

Only KIR and HLA ligands significantly associated with TTR and/or OS are shown. p values were

determined by the log-rank test.

Of 61 patients, 15 were carriers of the KIR2DS5 gene and 19 were homozygous for HLA-C1.

KIR2DS5 gene carriers presented better TTR and OS. Homozygous expression of HLA-C1 was

associated with better OS. Among 48 patients positive for HLA-Bw4, at least one copy of the I80 or

T80 variants was present in 31 and 23 patients, respectively. The I80 variant (either one or two

copies) was associated with better OS, while opposite behavior was shown by the T80 variant.

Figure 2. Effect of HLA-C1 and compound KIR/HLA genotypes on TTR and OS.

A: Homozygous HLA-C1 (HLA-C1C1; 2 copies, detected in 19 patients) was associated with

significantly longer OS compared to heterozygous HLA-C1 (HLA-C1C2; 1 copy, detected in 29

patients) and with significantly longer TTR and OS compared to homozygous HLA-C2 (HLA-

C2C2; 0 copies, detected in 13 patients). Survival curves are referred to each of the 3 possible

genotypes (HLA-C1 homozygous, heterozygous or null), while bar panels show the progressive

effect of HLA-C1 copy number on median TTR and OS. B: Effect of the compound KIR2DL2-

KIR2DS2/HLA-C1 and KIR3DS1/HLA-Bw4T80 genotypes on TTR and OS, respectively.

KIR2DL2 was detected in 34 patients, 2 of whom homozygous, and 26 also positive for HLA-C1

(homozygous or heterozygous); KIR2DS2 was detected in 33 patients, 25 also positive for HLA-C1

(homozygous or heterozygous); KIR3DS1 (homozygous or heterozygous) was detected in 18

patients, 9 of whom also positive for HLA-Bw4T80 (all heterozygous, either Bw4T80/BwI80 or

Bw4T80/Bw6).




18

Figure 3. Analysis of NK-cell phenotype and cytotoxicity.

Upper panel: left, Kaplan-Meier curves of TTR and OS of HCC patients with different %CD56dim

NK-cells. The cut-off value between high or low CD56dim% was determined by ROC curve

analysis. The area under the ROC curve (AUC) for TTR was 0.73 (0.56-0.90), and a cut-off value

of 88.95% CD56dim showed 75% sensitivity and 76.2% specificity. The AUC for OS was 0.77

(0.66-0.96), with 72.2% sensitivity and 78.6% specificity for a cut-off of 89.25% CD56dim; right,

cytotoxic potential measured by upregulation of CD107a in CD3-CD56+ cells upon stimulation

(mean±standard error) in subjects with or without HLA-C1/C2 or combined KIR2DL2/DL3-C1.

Lower panel: Flow cytometry analysis of NK cells cytotoxic potential is shown for a representative

patient (RTA14), positive for HLA-C1C2 and HLA-Bw4(T80)/Bw6. The KIR genotype of patient

RTA14 is shown in Supplementary Figure S1. The two plots on the left show gating of NK cells

after overnight PBMCs activation with IL-15, while the remaining plots show staining for CD107a

on gated NK-cells (CD3-CD56+) after incubation with or without K562 target cells.




19

References

1. Moretta A, Tambussi G, Bottino C, Tripodi G, Merli A, Ciccone E, et al. A novel surface

antigen expressed by a subset of human CD3- CD16+ natural killer cells. Role in cell

activation and regulation of cytolytic function. J Exp Med 1990;171:695-714.

2. Parham P. MHC class I molecules and KIRs in human history, health and survival. Nat Rev

Immunol 2005;5:201-14.

3. Colonna M, Borsellino G, Falco M, Ferrara GB, Strominger JL. HLA-C is the inhibitory

ligand that determines dominant resistance to lysis by NK1- and NK2-specific natural killer

cells. Proc Natl Acad Sci USA 1993;90: 12000–4.

4. Wagtmann N, Rajagopalan S, Winter CC, Peruzzi M, Long EO. Killer cell inhibitory

receptors specific for HLA-C and HLA-B identified by direct binding and by functional

transfer. Immunity 1995;3:801–9.

5. Moesta AK, Norman PJ, Yawata M, Yawata N, Gleimer M, Parham P. Synergistic

polymorphism at two positions distal to the ligand-binding site makes KIR2DL2 a stronger

receptor for HLA-C than KIR2DL3. J Immunol 2008;180:3969–79.

6. Gumperz JE, Litwin V, Phillips JH, Lanier LL, Parham P. The Bw4 public epitope of HLA-

B molecule confers reactivity with natural killer cell clones that express NKB1, a putative

HLA receptor. J Exp Med 1995;181:1133-44.

7. Stewart CA, Laugier-Anfossi F, Vély 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;102:13224-9.

8. Jamil KM, Khakoo SI. KIR/HLA interactions and pathogen immunity. J Biomed Biotechnol

2011;2011:298348.

9. Khakoo SI, Thio CL, Martin MP, Brooks CR, Gao X, Astemborski J, et al. HLA and NK cell

inhibitory receptor genes in resolving hepatitis C virus infection. Science 2004;305:872-4.




20

10. Romero V, Azocar J, Zúñiga J, Clavijo OP, Terreros D, Gu X, et al. Interaction of NK

inhibitory receptor genes with HLA-C and MHC class II alleles in hepatitis C virus infection

outcome. Mol Immunol 2008;45:2429-36.

11. Knapp S, Warshow U, Hegazy D, Brackenbury L, Guha IN, Fowell A, et al. Consistent

beneficial effects of killer cell immunoglobulin-like receptor 2DL3 and group 1 human

leukocyte antigen-C following exposure to hepatitis C virus. Hepatology 2010;51:1168-75.

12. Dring MM, Morrison MH, McSharry BP, Guinan KJ, Hagan R; Irish HCV Research

Consortium, et al. Innate immune genes synergize to predict increased risk of chronic disease

in hepatitis C virus infection. Proc Natl Acad Sci USA 2011;108:5736-41.

13. Amadei B, Urbani S, Cazaly A, Fisicaro P, Zerbini A, Ahmed P, et al. Activation of natural

killer cells during acute infection with hepatitis C virus. Gastroenterology 2010;138:1536-45.

14. Vidal-Castiñeira JR, López-Vázquez A, Díaz-Peña R, Alonso-Arias R, Martínez-Borra J,

Pérez R, et al. Effect of killer immunoglobulin-like receptors in the response to combined

treatment in patients with chronic hepatitis C virus infection. J Virol 2010;84:475-81.

15. Suppiah V, Gaudieri S, Armstrong NJ, O'Connor KS, Berg T, Weltman M, et al. IL28B,

HLA-C, and KIR variants additively predict response to therapy in chronic hepatitis C virus

infection in an European cohort: a cross-sectional study. PLoS Med 2011;8:e1001092.

16. Cai L, Zhang Z, Zhou L, Wang H, Fu J, Zhang S, et al. Functional impairment in circulating

and intrahepatic NK cells and relative mechanism in hepatocellular carcinoma patients. Clin

Immunol 2008;129;428-37.

17. López-Vázquez A, Rodrigo L, Martínez-Borra J, Pérez R, Rodríguez M, Fdez-Morera JL, et

al. Protective effect of the HLA-Bw4I80 epitope and the killer cell immunoglobulin-like

receptor 3DS1 gene against the development of hepatocellular carcinoma in patients with

hepatitis C virus infection. J Infect Dis 2005;192:162-5.

18. Alves LGT, Rajalingam R, Canavez F. A novel real-time PCR method for KIR genotyping.

Tissue Antigens 2008;73:188-91.




21

19. Gonzalez-Galarza FF, Christmas S, Middleton D, Jones AR. Allele frequency net: a database

and online repository for immune gene frequencies in worldwide populations. Nucleic Acids

Res 2011;39:D913-9.

20. Bontadini A, Testi M, Cuccia MC, Martinetti M, Carcassi C, Chiesa A, et al. Distribution of

killer cell immunoglobulin-like receptors genes in the Italian Caucasian population. J Transl

Med 2006;4:44.

21. Cella M, Longo A, Ferrara GB, Strominger JL, Colonna M. NK3-specific natural killer cells

are selectively inhibited by Bw4-positive HLA alleles with isoleucine 80. J Exp Med

1994;180:1235–42.

22. Carrington M, Wang S, Martin MP, Gao X, Schiffman M, Cheng J, et al. Hierarchy of

resistance to cervical neoplasia mediated by combinations of killer immunoglobulin-like

receptor and human leukocyte antigen loci. J Exp Med 2005;201:1069-75.

23. Kumar V, Kato N, Urabe Y, Takahashi A, Muroyama R, Hosono N, et al. Genome-wide

association study identifies a susceptibility locus for HCV-induced hepatocellular carcinoma.

Nat Genet 2011;43:455-8.

24. Moretta L, Bottino C, Pende D, Castriconi R, Mingari MC, Moretta A. Surface NK receptors

and their ligands on tumor cells. Semin Immunol 2006;18:151–8.

25. Fu J, Xu D, Liu Z, Shi M, Zhao P, Fu B, et al. Increased regulatory T-cells correlate with

CD8 T-cell impairment and poor survival in hepatocellular carcinoma patients.

Gastroenterology 2007;132: 2328–39.

26. Villanueva A, Hoshida Y, Battiston C, Tovar V, Sia D, Alsinet C, et al. Combining clinical,

pathology, and gene expression data to predict recurrence of hepatocellular carcinoma.

Gastroenterology 2011;140:1501-12.

27. Cariani E, Pilli M, Zerbini A, Rota C, Olivani A, Pelosi G, et al. Immunological and

molecular correlates of disease recurrence after liver resection for hepatocellular carcinoma.

PLoS One 2012;7:e32493.




22

28. Anfossi N, André P, Guia S, Falk CS, Roetynck S, Stewart CA, et al. Human NK cell

education by inhibitory receptors for MHC class I. Immunity 2006;25:331-42.

29. Kim S, Poursine-Laurent J, Truscott SM, Lybarger L, Song YJ, Yang L, et al. Licensing of

natural killer cells by host major histocompatibility complex class I molecules. Nature

2005;436:709-13.

30. Brodin P, Kärre K, Höglund P. NK cell education: not an on-off switch but a tunable

rheostat. Trends Immunol 2009;30:143-9.

31. Lanier LL. NK cell recognition. Ann Rev Immunol 2005;23:225-74.

32. Martin MP, Gao X, Lee JH, Nelson GW, Detels R, Goedert JJ, et al. Epistatic interaction

between KIR3DS1 and HLA-B delays the progression to AIDS. Nat Genet 2002;31:429-34.

33. Arnheim L, Dillner J, Sanjeevi CB. A population-based cohort study of KIR genes and

genotypes in relation to cervical intraepithelial neoplasia. Tissue Antigens 2005;65:252-9.

34. Butsch Kovacic M, Martin M, Gao X, Fuksenko T, Chen CJ, Cheng YJ, et al. Variation of

the killer cell immunoglobulin-like receptors and HLA-C genes in nasopharyngeal

carcinoma. Cancer Epidemiol Biomarkers Prev 2005;14:2673-7.

35. Pan N, Jiang W, Sun H, Miao F, Qiu J, Jin H, et al. KIR and HLA loci are associated with

hepatocellular carcinoma development in patients with hepatitis B virus infection: a case-

control study. PLoS One 2011;6:e25682.

36. Alici E. IPH-2101, a fully human anti-NK-cell inhibitory receptor mAb for the potential

treatment of hematological cancers. Curr Opin Mol Ther 2010;12:724-33.




1

Total

Surgical

resection

(SR)

Radiofrequency

Thermal Ablation

(RTA)

p value

(SR vs RTA)

Patients #

61

28

33

NA

Gender (M/F)

33/28 19/9 14/19 NS

Mean age at diagnosis ±

SD (years)

70.2±7.99 67.9±8.97 72.15±6.57 0.037

Child A/B

55/6 28/0 27/6 0.027

Median nodules # (range)

1 (1-5) 1 (1-5) 1 (1-3) NS

Mean nodules size ± SD

(mm)

42.3±23.6 58±25.7 29.9±11.7 <0.0001

Median TTR (months)

17

34

9

0.0002

Median OS (months)

43 84 41 0.05

Table 1. Clinical characteristics of patients.

SR: surgical resection; RTA: radiofrequency thermal ablation; NA: not applicable; NS: not

significant; M: males; F: females; SD: standard deviation.




2

TTR

OS

p value

HR (95% CI)

p value

HR (95% CI) Gender (M/F)

0.15

0.65 (0.36-1.17)

0.014

0.43 (0.22-0.84)

Age at diagnosis* 0.077 1.70 (0.94-3.05) 0.064 1.86 (0.96-3.61) Child (A vs B) 0.048 0.28 (0.08-0.99) 0.38 0.51 (0.11-2.31)

Treatment (SR vs RTA) <0.001 0.496 (0.246-1.00) 0.0493 0.312 (1.00-4.06)

KIR2DS5 0.024 0.5 (0.27-0.91) 0.029 0.47 (0.24-0.93) KIR2DS1

0.06

0.57 (0.32-1.02)

0.11

0.58 (0.29-1.14)

HLA-Bw4I80

0.72

0.90 (0.51-1.59)

0.0499

0.43 (0.18-1)

HLA-Bw4T80

0.57

1.20 (0.64-2.23)

0.008

2.93 (1.33-6.43)

HLA-C1

0.057

0.44 (0.19-1.03)

0.13

0.50 (0.2-1.24)

HLA-C2

0.093

1.65 (0.92-2.98)

0.011

2.36 (1.22-4.54)

KIR2DL2+/C1+ vs C1-

0.027

0.29 (0.09-0.86)

0.56

0.72 (0.24-2.19)

KIR2DL3+/C1+ vs C1-

0.073

0.42 (0.16-1.08)

0.310

0.61 (0.23-1.59)

KIR2DS2+/C1+ vs C1- KIR3DS1+/I80+ vs I80-

0.026

0.91

0.28 (0.09-0.86)

1.08 (0.27-4.36)

0.5897

0.079

0.7371 (0.24-2.23)

0.17 (0.02-1.23)

KIR3DS1+/T80+ vsT80-

0.81

1.17 (0.32-4.30)

0.020

5.64 (1.32-24.16)

Table 2. TTR and OS of HCC patients according to clinical characteristics, KIRs, KIR ligands, and

compound KIR-HLA genotypes. Only parameters presenting p<0.1 in at least one analysis (TTR

and/or OS) are shown.

* Age ≥ vs < median (71 years); HR: hazard ratio; CI: confidence interval.




3

TTR

OS

p value HR (95% CI) p value HR (95% CI)

Gender (M vs F) NA NA 0.342 0.70 (0.33-1.46)

Age at diagnosis

( ≥71 vs <71 years) 0.106 1.67 (0.89-3.12) 0.114 1.85 (0.86-3.96)

Child (A vs B) 0.106 0.43 (0.160-1.19) NA NA

Treatment (SR vs RTA) 0.021 0.43 (0.21-0.88) 0.992 0.99 (0.42-2.34)

KIR2DS5 (pos vs neg) 0.664 0.74 (0.19-2.91) 0.047 0.408 (0.17-0.99)

KIR2DS1 (pos vs neg) 0.698 0.78 (0.22-2.79) NA NA

HLA-C1 (pos vs neg) 0.023 0.40 (0.18-0.88) NA NA

HLA-C2 (pos vs neg) 0.961 1.02 (0.50-2.09) 0.067 2.17 (0.99-4.98)

Table 3. Multivariate analysis of TTR and OS.

NA: not applicable (p>0.10 in the univariate analysis).




urvi

val

60

80

100KIR2DS5-KIR2DS5+

p=0.0293

surv

ival

60

80

100 KIR2DS5-KIR2DS5+

p=0.0244

Fig. 1

Perc

ent s

u

0 50 100 1500

20

40

Perc

ent s

0 50 100 1500

20

40

OS months0 50 100 150

al 80

100 HLA-C1C1HLA-C1C2/HLA-C2C2

TTR months0 50 100 150

Perc

ent s

urvi

va

20

40

60

80p=0.0105

100 HLA-Bw4 I80- 100 HLA-Bw4 T80-

OS months0 50 100 150

0

cent

sur

viva

l

40

60

80

100HLA-Bw4 I80+

p=0.0499

cent

sur

viva

l

40

60

80

100HLA-Bw4 T80+

p=0.0075

OS months

Perc

0 50 100 1500

20

OS months

Perc

0 50 100 1500

20

Research.

on February 14, 2019. ©

2013 Am

erican Association for C

ancerclincancerres.aacrjournals.org

Dow

nloaded from

Author m

anuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Author M

anuscript Published O

nlineFirst on A

ugust 12, 2013; DO

I: 10.1158/1078-0432.CC

R-13-0986


surv

ival

60

80

100HLA-C2C2HLA-C1C2HLA-C1C1

urvi

val

60

80

100

HLA-C2C2HLA-C1C2HLA-C1C1

p=0.034

Fig. 2

Perc

ent

0 50 100 1500

20

40 p=0.122

Perc

ent s

u

0 50 100 1500

20

40

60

HLA-C1C1

TTR months

HLA-C1C1

p=0.022

OS months

0 50 100

150

HLA-C2C2

HLA-C1C2

p=0.037

0 50 100

150

HLA-C2C2

HLA-C1C2

p=0.018

TTR monthsmedian (range)

OS monthsmedian (range)

100 KIR2DL2+/C1- KIR3DS1+/Bw4T80-

A

100 KIR2DS2+/C1-

cent

sur

viva

l

40

60

80

100KIR2DL2+/C1+

p=0.0265

ent s

urvi

val

40

60

80

100 KIR3DS1+/Bw4T80+

p=0.0197

cent

sur

viva

l

40

60

80

100 S /CKIR2DS2+/C1+

p=0.0262

TTR months

Perc

0 50 100 1500

20

OS months

Perc

0 50 100 1500

20

BTTR months

Perc

0 50 100 1500

20

Research.

on February 14, 2019. ©

2013 Am

erican Association for C

ancerclincancerres.aacrjournals.org

Dow

nloaded from

Author m

anuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Author M

anuscript Published O

nlineFirst on A

ugust 12, 2013; DO

I: 10.1158/1078-0432.CC

R-13-0986


TTR months

Perc

en

t su

rviv

al

0 50 100 150 0

20

40

60

80

100

p=0.0002

CD56DIM low

CD56DIM high

OS months

Perc

en

t su

rviv

al

0 50 100 150 0

20

40

60

80

100

p=0.0175

CD56DIM low

CD56DIM high

% C

D5

6/C

D1

07

a+

C1-

C1+ C

2-C2+

KIR

2DL2+

/C1-

KIR

2DL2+

/C1+

KIR

2DL3+

/C1-

KIR

2DL3+

/C1+

0

10

20

30

40

50

p=0.035 p=0.0075 p=0.0086

FSC CD3 CD56

SSC

CD

56

CD

10

7a

CD

10

7a

CD56

Fig. 3




Published OnlineFirst August 12, 2013.Clin Cancer Res Elisabetta Cariani, Massimo Pilli, Alessandro Zerbini, et al. virus-related hepatocellular carcinomareceptor genes as outcome predictors of hepatitis C Human Leukocyte Antigen and Killer Immunoglobulin-like

Updated version

10.1158/1078-0432.CCR-13-0986doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2013/08/12/1078-0432.CCR-13-0986.DC1

Access the most recent supplemental material at:

Manuscript

Authoredited. Author manuscripts have been peer reviewed and accepted for publication but have not yet been

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/early/2013/08/10/1078-0432.CCR-13-0986To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-13-0986

http://clincancerres.aacrjournals.org/content/suppl/2013/08/12/1078-0432.CCR-13-0986.DC1

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/early/2013/08/10/1078-0432.CCR-13-0986


Human Leukocyte Antigen and Killer Immunoglobulin-like...

Documents

Transcript of Human Leukocyte Antigen and Killer Immunoglobulin-like...