WO2012072595A1

WO2012072595A1 - Platelet factor 4 variant for the prognosis of cardiovascular outcome in patients having heart disease

Info

Publication number: WO2012072595A1
Application number: PCT/EP2011/071207
Authority: WO
Inventors: Johan De Sutter; Jo Van Damme; Sofie Struyf; Marc De Buyzere; Jan PHILIPPÉ; Nico Van De Veire
Original assignee: Universiteit Gent; Katholieke Universiteit Leuven
Priority date: 2010-11-30
Filing date: 2011-11-28
Publication date: 2012-06-07

Abstract

The present invention relates to markers for the prognosis of the likely outcome of cardiovascular disease in patients. More particularly, the present invention discloses that low levels of the chemokinedenominated as 'Platelet factor 4 variant' (PF4var) in serum of patients having cardiovascular disease and preserved left ventricular function are predictive for a negative outcome. Hence, the invention relates to methods for the prognosis of cardiovascular disease using PF4var and kits to perform said methods.

Description

Platelet factor 4 variant for the prognosis of cardiovascular outcome

in patients having heart disease

Technical field of the invention

The present invention relates to markers for the prognosis of the likely outcome of heart disease in patients. More particularly, the present invention discloses that low levels of the chemokine denominated as 'Platelet factor 4 variant' (PF4var) in serum of patients having coronary artery disease and preserved left ventricular function are predictive for a negative outcome in the future. Hence, the invention relates to methods for the prognosis of heart disease using PF4var and kits to perform said methods.

Background art

Heart disease or cardiopathy comprises a variety of different diseases affecting the heart such as coronary heart disease, coronary artery disease, cardiomyopathy, cardiovascular disease, ischemic heart disease, heart failure, hypertensive heart disease, inflammatory heart disease and valvular heart disease. Coronary heart disease refers to the failure of coronary circulation to supply adequate circulation to cardiac muscle and surrounding tissue. It is already the most common form of disease affecting the heart and an important cause of premature death in Europe, the Baltic states, Russia, North and South America, Australia and New Zealand. It has been predicted that all regions of the world will be affected by 2020. Hence, biomarkers which can help with the diagnosis and prognosis of heart disease are of utmost importance.

Biomarkers currently most widely used in detection of myocardial damage are the MB subtype of the enzyme creatine kinase and cardiac troponins T and I. For example, the cardiac troponins T and I are released within 4-6 hours of an attack of myocardial infarction and remain elevated for up to 2 weeks. Hence, they are now the preferred markers for assessing myocardial damage. Elevated troponins in the setting of chest pain may accurately predict a high likelihood of a myocardial infarction in the near future. New markers such as glycogen phosphorylase isoenzyme BB are under investigation. Another marker such as proBrain natriuretic peptide (NT-proBNP) has been approved as a marker for acute congestive heart failure.

Platelet factor 4 (PF4) or CXCL4, the first discovered chemokine, is selectively released from stimulated platelets and has atypical biological properties in that it is only a weak leukocyte chemoattractant compared to other chemokines. On the other hand, PF4 is reportedly implicated in many biological processes, such as inhibition of hematopoiesis, platelet coagulation, and activation of various myeloid and lymphoid leukocyte types [1]. Further, PF4/CXCL4 is implicated in pathological processes, such as atherosclerosis and heparin-induced thrombocytopenia [4]. Indeed, platelet- derived chemokines, including PF4, influence several aspects of vascular biology. PF4 enhances monocyte arrest on endothelial cells induced by ANTES/CCL5 (i.e. another platelet-derived chemokine) and boosts atherogenesis (Koenen et al., Nat Med 2009). Indeed, atherosclerotic lesions were smaller in PF4^7", compared to wild type mice (Sachais et al., Thromb Haemost 2007). PF4 has been demonstrated in human atherosclerotic plaques and PF4 takes part in the metabolism of atherogenic lipids (von Hundelshausen et al. Thromb Haemost 2007). In patients with coronary artery disease (CAD) or peripheral vascular disease altered plasma concentrations of PF4 were measured. Over a narrow range of concentration PF4 itself promotes thrombosis, however at low or very high doses the thrombotic effect of PF4 cannot be observed. Lambert et al. proposed a model for this apparent contradictory results (21).

The gene of 'platelet factor 4 variant' (PF4var) or CXCL4L1, also denominated as PF4alt/PF4varl, has been described as a non-allelic PF4 variant of PF4 (Green et al. 1989 Mol Cell Biol:1445; Eisman et al. 1990 Blood: 336). WO90/08824 describes the use of PF4var protein as an immunostimulant in an immunosuppressed mouse model. WO94/07524 describes the use of PF4var to treat arthritis and other inflammatory diseases. US 2009/0011981 discloses the anti-angiogenic activity of PF4var. A most striking activity of PF4var, shared with PF4, is the inhibition of endothelial cell proliferation and migration [2]. Angiogenesis induced by angiogenic chemokines such as interleukin-8, fibroblast growth factor (FGF) or vascular endothelial growth factor (VEGF) was significantly reduced by PF4var and by PF4. In particular, PF4var was found to be a more potent angiostatic chemokine than PF4 with stronger antitumoral activity in various animal models [3]. Nothing is known about PF4var in relation to cardiac diseases.

The present invention discloses that, although it is known that PF4 and PF4var are similar in some aspects, low levels of PF4var -but not of PF4- in samples from patients having heart disease surprisingly correlate with a negative prognosis. Hence, the present invention relates to PF4var as a new biomarker for use in the prognosis of heart disease. Brief description of figures

Figure 1: Kaplan-Meier curves for patients with levels of NT-proBNP lower and above the median value of 164 pg/ml (figure 1A) and Kaplan-Meier curves for patients with levels of PF4var/CXCL4Ll lower and above the median value of 10 ng/ml (figure IB).

Figure 2: Kaplan-Meier curves showing freedom from the primary outcome measure (death due to cardiovascular causes, non-fatal myocardial infarction, non-fatal stroke or hospitalization for congestive heart failure) according to 4 groups: group 1: NT-proBNP < median and PF4var/CXCL4Ll > median (n=64); group 2: NT-proBNP < median and PF-4var/CXCL4Ll < median (n=42); group 3: NT- proBNP > median and PF-4var/CXCL4Ll > median (n=43); group 4: NT-proBNP > median and PF4var/CXCL4Ll < median (n=56).

Figure 3: Kaplan-Meier curves showing freedom from the secondary outcome measure (death due to cardiovascular causes, non-fatal myocardial infarction, percutaneous coronary intervention, coronary artery bypass grafting or hospitalization for congestive heart failure) according to 4 groups: group 1: NT-proBNP < median and PF4var/CXCL4Ll > median (n=64); group 2: NT-proBNP < median and PF4var/CXCL4Ll < median (n=42); group 3: NT-proBNP > median and PF4var/CXCL4Ll > median (n=43); group 4: NT-proBNP > median and PF4var/CXCL4Ll < median (n=56).

Detailed description of invention

The present invention discloses that low levels of PF4var -but not of PF4- in samples from patients having heart disease correlate with a negative prognosis. Hence the present invention relates to the in vitro usage of the chemokine 'Platelet factor 4 variant' (PF4var) comprising the amino acid sequence as depicted by SEQ ID N°l or a fragment thereof comprising at least the amino acid sequence LLYKKIIKEH (= SEQ ID N°2) as a marker for the prognosis of cardiovascular outcome in patients having heart disease.

The term 'PF4var' in the present invention refers to the natural, synthetic or recombinant versions of PF4var. The PF4var protein is also known as PF-4var, PF4alt, PF4V1, SCYB4V1 or CXCL4L1 (OM IM number 173461, accession numbers P10720 and M26167 at Swiss-Prot and Genbank databases, for the protein and genomic DNA sequence, respectively). The mature PF4var protein is obtained by processing of a longer propeptide. The 70 amino acid-, mature protein has the following sequence using the 1 letter amino acid code: EAEEDGDLQC LCVKTTSQV PRHITSLEVI KAGPHCPTAQ LIATLKNGRK ICLDLQALLY KKIIKEHLES (= SEQ ID N°l). The numbering of the amino acids in this sequence is used for referral throughout the present application. The latter version is the most abundant form of PF4var in platelet preparations as is indicated in US 2009/0011981. Also longer sequences (SEQ ID N°l extended at its aminoterminus by FA A or SEQ ID N°l extended at its aminoterminus by A) and partially processed versions of the PF4var protein, e.g. missing the most NH₂-terminal E in the SEQ ID N°l, or some COOH-terminal residues (e.g. SEQ ID N°l missing the COOH-terminal S) occur in lower abundance in platelet preparations and are also part of the present invention. Hence, and for example, a chemokine consisting of about 80 amino acids and comprising SEQ ID N° 1, or, a chemokine fragment of about 30 amino acids and comprising the sequence LLYKKIIKEH (SEQ ID N° 2, see further) are chemokines fragments which are part of the present invention. Furthermore, the mature protein of PF4var differs in only 3 amino acids located in the COOH-terminus from PF4: namely PF4var has a leucine (L) on position 58 instead of a proline (P) in PF4, a glutamic acid (E) on position 66 instead of a lysine (K) and a histidine (H) on position 67 instead of a leucine (L).

The term 'fragment thereof comprising at least the amino acid sequence LLYKKIIKEH' as used herein refers to a protein/peptide which comprises at least 10 amino acids (= at least the 10 consecutive amino acids LLYKKIIKEH corresponding to the position 58 (L) up to 67 (H) of SEQ ID N° 1), which is shorter than the 70 amino acids of SEQ ID N°l, and which differentiates PF4var from PF4 as the amino acid region LLYKKIIKEH comprises the positions 58 (L), 66 (E) and 67 (H). Hence, said fragments consist, for example, of 10, 11, 12, 13, 14..., 20,..., 30,..., 40,..., 50,..., 60,..., 65, 66, 67, 68 or 69 amino acids.

With the term 'marker' or 'biomarker' is meant a characteristic that is objectively measured and evaluated as an indicator of normal biologic processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention. Hence, the biomarker PF4var of the present invention can be used, among other uses, to: 1) evaluate the prognosis of heart disease and in particular to diagnose coronary heart disease with preserved left ventricular function, which encompasses predictions about the likely course of disease or disease progression, particularly with respect to the likelihood of nonfatal myocardial infarction, nonfatal stroke, hospitalization for congestive heart failure and death due to cardiovascular causes; 2) therapeutically stratify patients with heart disease, in particular coronary heart disease with preserved left ventricular function, in order to decide which therapy should be given to said patient; and 3) monitor disease progression once a particular therapy has been administered to said patients. Potential examples of such therapies include, but are not limited to, anti-platelet drugs, anti-coagulant drugs and thrombolytic drugs. The terms 'negative prognosis' and 'cardiovascular outcome' refer to a nonfatal myocardial infarction, nonfatal stroke, hospitalization for congestive heart failure or death due to cardiovascular causes of a patient within a certain time period (for example 1, 2 to 3 years) after determining PF4var levels in a sample from said patient.

The term 'heart disease' refers to coronary heart disease, coronary artery disease, acute aortic dissection, cardiomyopathy, cardiovascular disease, ischemic heart disease, heart failure, hypertensive heart disease, inflammatory heart disease and valvular heart disease. In particular, the term 'heart disease' refers to coronary heart disease with preserved left ventricular function. Preserved left ventricular function means that the left ventricular ejection fraction is equal or higher than 50% of the total ejection fraction before onset of disease. The term 'coronary heart disease' or 'coronary artery disease' (CAD) refers to at least one of the following clinical observations: a previous history (> 6 months) of acute myocardial infarction (AM I), percutaneous coronary intervention (PCI), coronary artery bypass grafting (CABG), or documented CAD on coronary angiography (> 70% stenosis).

Consequently, the present invention relates to the in vitro use of a chemokine, particularly PF4var, as a marker for the prognosis of cardiovascular outcome in a patient having heart disease as described above and particularly wherein said heart disease is coronary artery disease with preserved left ventricular function.

The chemokine of the present invention further relates to the usage of a chemokine, particularly PF4var, which is used together with another marker for the prognosis of cardiovascular outcome in patients having heart disease. Non-limiting examples of such markers are the enzyme creatine kinase, cardiac troponins T and I, glycogen phosphorylase isoenzyme BB, heart fatty acid binding protein, D-dimer, C-reactive protein, matrix metalloproteinase-9 and, preferably, N-terminal proBrain natriuretic peptide (NT-proBNP).

For example, the present invention discloses significant correlations between PF4var levels and age, creatinine, NT-proBNP and circulating platelet number.

The present invention specifically relates to the usage of a chemokine as described above wherein said other marker is N-terminal proBrain natriuretic peptide (NT-proBNP).

The present invention also concerns an in vitro method to prognose-, or to a method to collect data for the prognosis of-, the cardiovascular outcome in a patient having heart disease comprising the steps: 1) obtaining a sample from said patient, and 2) determining the level of PF4var having the amino acid sequence as depicted by SEQ. ID N°l or a fragment thereof comprising at least the amino acid sequence LLYKKIIKEH. In a particular embodiment of the present invention, said in vitro method comprises the steps: 1) obtaining a sample from said patient, and 2) determining the level of PF4var having the amino acid sequence as depicted by SEQ ID N°l or a fragment thereof comprising at least the amino acid sequence LLYKKIIKEH wherein the determined level of PF4var leads to a negative prognosis for said patient if the level is lower than the median PF4var level of the population to which said patient belongs. In other words, the prognosis of said patient will be negative if the level of PF4var having the amino acid sequence as depicted by SEQ ID N°l or a fragment thereof comprising at least the amino acid sequence LLYKKIIKEH is lower than a specifically chosen - depending on the preferred specificity and/or sensitivity of the prognosis- cut-off value. The latter value can be the median level of PF4var having the amino acid sequence as depicted by SEQ ID N°l or a fragment thereof comprising at least the amino acid sequence LLYKKIIKEH in serum as measured in a specific study population. For example, said median level of the study population of the present invention is 10 ng/ml (interquartile range 8-16 ng/ml)but can be also equal to 7, 8, 9, 11, 12, 13 or 14 ng/ml in another study population. Moreover the median level of a particular study population of the present invention might be higher for women (14 [interquartile range 10-22] ng/ml) compared to the median level for men (10 [interquartile range 7-15] ng/ml). The higher the chosen cut-off value is, the higher the sensitivity of the prognosis will be. Another example of a cut-off value that can be chosen is the level of PF4var in healthy individuals. The present invention discloses that -in a specific study population- the mean value of PF4var in healthy individuals was 14.5±7.3 (range 5-33) without significant differences between men and women (13.5±7.4 vs 15.8±7.1, respectively).

The concentrations (such as mg/ml, ng/ml, pg/ml, etc) as depicted in this invention always refer to the concentration of a particular compound/protein/peptide in serum.

The term ' a sample from said patient' includes, but is not limited to a tissue or a biofluid such as blood, serum, plasma lymph or any other bodily secretion or derivative thereof. Methods for collecting or obtaining various samples are well known in the art. The present invention specifically relates to a method as described above wherein said sample is a blood sample.

The terms 'determining the level of PF4var' refers to measuring or determining a low or decreased level of PF4var in a patient sample having a negative prognosis compared to the cut-off value, such as the median value as indicated above, of the population to which said patient belongs. For example, median PF-4var of the present study was 10 ng/ml (interquartile range 8-16 ng/ml) and independent determinants of PF-4var levels were age, gender and circulating platelet number. Patients who experienced cardiac events (n=20) during follow-up showed lower levels of PF- 4var/CXCL4Ll (8.5 [5.3-10] ng/ml versus 12 [8-16] ng/ml, p=0.033).

Said 'level' might further depend on which level of PF4var is measured and how this level is measured. In a particular embodiment, the present invention relates to an in vitro method to prognose the cardiovascular outcome wherein the level of PF4var is determined by measuring the level or amount of PF4var protein. For example, the present invention specifically relates to a sandwich ELISA for the chemokine PF4var as described by Vandercappellen et al. (2007; J. Leuk Biol: 1519). Using the latter ELISA, low levels of PF4var (negative prognosis) are about 8.5 ng PF4var per ml serum whereas high levels (positive prognosis) are about 12 ng PF4var per ml serum. Alternatively said PF4var level can be determined by measuring the level of nucleic acids, such as m NA expression, of PF4var. Measuring proteins and nucleic acid levels (such as mRNA levels) are well known in the art and can be undertaken by any method known in the art including but not limited to Western blots, Northern blots, Southern blots, ELISA, multiplex-ELISA systems such as cytometric bead arrays, immunoprecipitation, immunofluorescense, flow cytometry, immunohistochemistry, nucleic acid hybridization techniques, nucleic acid reverse transcription methods, and nucleic acid amplification methods such as qPCR. The latter techniques are, for example, described in detail in US 2007/0218512. Different types of mass spectrometry are other well known methods in the art which can be employed to detect PF4var levels. In particular embodiments, expression of a biomarker is detected on a protein level using antibodies that are directed against specific biomarker proteins as described above. These antibodies can be used in various methods such as Western blot, ELISA, immunoprecipitation or immunohistochemistry.

Preferred embodiments of the method of the present invention as described above are methods wherein said PF4var protein level is determined via a sandwich-type ELISA and wherein said level of PF4var mRNA is determined via a quantitative RT-PCR assay (Verbeke et al. Hum Pathol 2010).

The present invention further relates to the use of a kit to prognose the cardiovascular outcome in patients having heart disease wherein said kit comprises reagents to perform an in vitro assay for determining the level of PF4var having the amino acid sequence as depicted by SEQ ID N°l, or a fragment thereof comprising at least the amino acid sequence LLYKKIIKEH, in a sample obtained from said patients.

The term 'kit' refers to any manufacture (e.g. a package or a container) comprising at least one reagent (e.g. an antibody, a nucleic acid probe, etc.) for performing an assay which specifically detects the level of PF4var. Positive and/or negative controls can be included in the kits to validate the activity and correct usage of reagents employed in accordance with the present invention. The design and use of controls is standard and well within the routine capabilities of those of ordinary skill in the art. The kit can be promoted, distributed, or sold as a unit for performing the methods or usages of the present invention. Additionally, the kits can contain a package insert describing the kit and methods/usages for its use.

The present invention particularly relates to the use of a kit as described above wherein said assay is a sandwich-type ELISA or a quantitative T-PC assay.

The phrase "specifically (or selectively) detects (or binds)" a protein or peptide, when referring to an antibody, refers to a binding reaction that is determinative of the presence of the protein in a heterogeneous population of proteins and other biologies. Thus, under designated immunoassay conditions, the specified antibodies bind to a particular protein at least two times the background and do not substantially bind in a significant amount to other proteins present in the sample. Specific binding to an antibody under such conditions may require an antibody that is selected for its specificity for a particular protein. For example, polyclonal antibodies raised to marker "X" from specific species such as rat, mouse, or human can be selected to obtain only those polyclonal antibodies that are specifically immunoreactive with marker "X" and not with other proteins, except for polymorphic variants and alleles of marker "X". This selection may be achieved by subtracting out antibodies that cross-react with marker "X" molecules from other species. A variety of immunoassay formats may be used to select antibodies specifically immunoreactive with a particular protein. For example, solid-phase ELISA immunoassays are routinely used to select antibodies specifically immunoreactive with a protein (see, e.g., Harlow & Lane, Antibodies, A Laboratory Manual (1988), for a description of immunoassay formats and conditions that can be used to determine specific immunoreactivity). Typically a specific or selective reaction will be at least twice background signal or noise and more typically more than 10 to 100 times background. "Immunoassay" is an assay that uses an antibody to specifically bind an antigen (e.g. a marker). The immunoassay is characterized by the use of specific binding properties of a particular antibody to isolate, target, and/or quantify the antigen.

The following non-limitative example is given in order to further illustrate the present invention. Example

METHODS

Study population

In order to obtain normal values for PF-4var, we evaluated 47 normal subjects. These individuals had no history of cardiovascular disease or diabetes, had no cardiac complaints and showed normal findings on a resting ECG and echocardiogram. For the CAD patients, we prospectively evaluated 205 consecutive patients with stable CAD between October 2002 and July 2005. The following clinical observations were considered as criteria for CAD: previous history (> 6 months) of acute myocardial infarction (AMI), percutaneous coronary intervention (PCI), coronary artery bypass grafting (CABG), or documented CAD on coronary angiography (> 70% stenosis). Patients with crescendo angina or angina at rest were excluded, as well as patients with recent (<6 months) acute coronary syndromes or cardiac revascularizations (17).

After overnight fasting, patients underwent a study protocol including venous blood sampling, echocardiography and evaluation of six-minutes walking distance. Echocardiography was performed with a VIVID-7 scanner (GE Vingmed Ultrasound, Horten, Norway). The left ventricular ejection fraction (LVEF) was measured in all patients using Simpson's method of discs. Patients with a LVEF > 50% were included in the present study. A six-minutes walking distance test was performed as measurement of exercise capacity. Routine blood measurements (blood cell count, creatinine) were performed according to an ISO 17025 Beltest accreditation.

Laboratory measurements

Serum concentration of NT-proBNP was measured on an Elecsys 2010 apparatus (Roche Diagnostics, Mannheim, Germany) with an automated electrochemiluminescence sandwich immunoassay. Soluble tumour necrosis factor receptors I and II (sTNFRI and II) were measured as parameters of inflammation by ELISA (BioSource, Nivelles, Belgium) with sensitivities of 0.05 and 0.1 ng/ml, respectively.

The CC chemokine RANTES/CCL5 was measured using the ELISA antibody pair distributed by R&D Systems (Abingdon, UK), whereas sandwich ELISAs for the CXC chemokines PF4/CXCL4 and PF4var/CXCL4Ll were developed as described (18). Antibodies against PF4 and PF4var were raised in rabbits as follows: immunization of rabbits was performed using a synthetic peptide of PF4var, PF4var^{47 70}, comprising the 24 C-terminal residues of the chemokine (NGRKICLDLQALLYKKIIKEHLES). For the first injection, the antigen (20 or 40 μg) was emulsified in 1.5 ml complete Freund's Adjuvant (FA). The rabbits were boosted at least four times at intervals of 4 weeks with 20 μg PF4var^{47 70} in 1.5 ml incomplete FA. Blood was collected by bleeding the rabbits at the ear vein 10 and 20 days after immunization. Antibodies were purified through protein G affinity chromatography (Pharmacia Biotech, Uppsala, Sweden). The specificity of the antibodies was verified by western blotting (Vandercappellen et al. J Leuk Biol 2007) and in an ELISA binding assay. ELISA plates were coated with antigen (PF4^{47 70}, PF4var^{47 70}, PF4 or PF4var; 20 ng/ml; 100 μΙ/cup). After saturation of the binding sites on the plate with casein, dilutions of the generated rabbit antisera were loaded. Immune complexes were detected with anti-rabbit-horse radish peroxidase and peroxidase activity was quantified by measuring the conversion of 3,3',5,5'-tetramethylbenzidine at 450 nm. The specificity of the purified antibodies was once more confirmed after development of a sandwich ELISA, using our rabbit polyclonal antibodies and a commercial monoclonal anti-PF4 (R&D Systems) (Vandercappellen et al. J Leuk Biol 2007).

Natural PF4/CXCL4 and recombinant PF4var/CXCL4Ll, purified to homogeneity in our laboratory, were used as standards. The PF4var/CXCL4Ll ELISA was specific in that PF4/CXCL4 was not detectable, whereas the PF4/CXCL4 ELISA measured both PF4/CXCL4 and PF4var/CXCL4Ll (18).

Follow-up

Patients were followed thereafter for the combined primary outcome measure of death due to cardiovascular causes, non-fatal myocardial infarction, non-fatal stroke or hospitalization for congestive heart failure. As secondary outcome measure the combination of MACE (death due to cardiovascular causes, non-fatal myocardial infarction, percutaneous coronary intervention or coronary artery bypass grafting) or hospitalization for congestive heart failure was evaluated. No patient was lost during the follow-up.

Ethics

This study protocol was approved by the ethical committee of the Ghent University Hospital (Belgium) and all patients gave informed consent.

Statistical analysis

Statistical analysis was performed using SPSS v 17.0 for Windows (SPSS Inc., 2008, Chicago, IL, USA). All continuous variables were tested for normality and log-transformed when skewed distributions were demonstrated. These are presented as mean ± standard deviation or median and interquartile range when appropriate. Differences between groups were analysed using ANOVA or Wilcoxon for continuous variables and chi-square for categorical variables. Spearman correlations were used to test the association of continuous variables with levels of PF4var/CXCL4Ll. Linear regression analysis was used to evaluate independent predictors of PF4var/CXCL4Ll levels. For outcome analysis, multivariate analysis was performed using Cox proportional hazards models incorporating factors with predictive significance on univariate analysis and ROC analysis and hypothesized important factors. Three models were constructed:

Model 1 clinical parameters (including age, gender, diabetes, creatinine, ejection fraction, treatment with betablockers, statins, antiplatelet drugs and ACE/ARB) with the addition of NT-proBNP.

Model 2 included the same clinical parameters as model 1, with the addition of PF4var/CXCL4Ll.

Model 3 included the same clinical parameters as model 1 with addition of NT-proBNP and PF4var/CXCL4Ll.

Cumulative event-free survival rates as a function over time were obtained by the Kaplan-Meier method. Differences in survival were analyzed by log-rank testing.

For all analyses the level of significance (p) was set at 0.05.

RESULTS

PF4var levels and determinants in healthy control subjects

Mean age of the normal subjects was 63±11 years (range 28-79 years) and 26 were men (55%). Mean LVEF was 63±7%, mean NT-proBNP 102±83 pg/ml and mean creatinine 0.89±0.17 mg/dl. Hypertension was present in 19 individuals (40%) which required treatment with ACE inhibitors or ARB in 9 (19%), beta-blockers in 10 (21%), diuretics in 8 (17%) and calcium antagonists in 5 subjects (11%). Statins were used in 10 (21%) and aspirine in 13 individuals (28%).

Mean value of PF4var was 14.5±7.3 (range 5-33) without significant differences between men and women (13.5±7.4 vs 15.8±7.1, p=0.19). No significant correlations were documented between PF4var levels and age, body mass index, LVEF, creatinine and circulating platelets. Only a weak correlation was present between PF4var and NT-proBNP (r=-0.31, p=0.04). Finally no significant differences in PF4var levels were noted in subjects with and without hypertension (13.9±6.7 vs 15.2±7.9, p=0.55) or aspirine use (16.2±8.2 vs 13.9±6.9, p=0.37).

Clinical characteristics of the study population

The clinical and laboratory characteristics of the total study population and of the subgroups according to the median PF4var/CXCL4Ll value of 10 ng/ml are shown in table 1. Patients with lower PF4var/CXCL4Ll levels were more frequently males, had a higher creatinine level and a tendency towards higher NT-proBNP levels. However, other clinical characteristics, including risk factors, previous cardiac history, medical treatment, markers of severity of heart failure and inflammation were comparable between both groups.

Determinants of PF4var/CXCL4Ll levels

Significant correlations were noted between PF4var/CXCL4Ll levels and age (r=-0.15, p=0.03), creatinine (r=-0.21, p=0.002), NT-proBNP (r=-0.21, p=0.003) and circulating platelet number (r=0.233, p=0.001). Median levels of PF4var/CXCL4Ll were significantly lower in men as compared to women (10 [7-15] ng/ml versus 14 [10-22] ng/ml, p<0.01). No significant correlations were noted between PF4var/CXCL4Ll and sTNFRI, sTNFRII, number of circulating white blood cells, lymphocytes, granulocytes or monocytes.

In multivariate regression analysis including age, gender, creatinine, NT-proBNP and circulating platelet number, levels of PF4var/CXCL4Ll were independently predicted by age (β = -0.145, p=0.028), gender (β = -0.239, p=0.001), and circulating platelet number (β = 0.218, p<0.001).

Relationship between PF4var/CXCL4Ll levels and levels of PF4/CXCL4 and RANTES/CCL5

As shown in table 1, levels of PF4/CXCL4 and RANTES/CCL5 were significantly lower in patients with lower PF4var/CXCL4Ll levels. A modest but significant correlation was noted between PF4var/CXCL4Ll levels and PF4/CXCL4 (r=0.29, p=0.001) but not with RANTES/CCL5 (r=0.125, p=0.074). In multivariate regression analysis including age, gender, creatinine, NT-proBNP and circulating platelet number, levels of PF4/CXCL4 were independently predicted by gender (β = 0.157, p=0.024), and circulating platelet number (β = 0.395, p<0.001). Levels of RANTES/CCL5 were only determined by circulating platelet number (β = 0.165, p=0.023).

Outcome

After a median follow-up of 2.5 years, 20 patients experienced the primary outcome measure. In patients with events, levels of NT-proBNP were higher than in patients without events (492 [267- 1387] pg/ml versus 257 [76-292] pg/ml, p<0.001), whereas levels of PF4var/CXCL4Ll were lower (8.5 [5.3-10] ng/ml versus 12 [8-16] ng/ml, p=0.033). In contrast, levels of PF4/CXCL4, RANTES/CCL5 and platelet counts were similar in patients with and without events. ROC analysis for events showed an area under the curve (AUC) of 0.82 (95% CI 0.73-0.90, p<0.001) for higher NT-proBNP levels and an AUC of 0.32 (95% CI 0.19-0.45, p=0.009) for lower PF4var/CXCL4Ll levels. No significant AUC was detected for PF4/CXCL4, RANTES/CCL5 or circulating platelets. Figure 1 shows the Kaplan-Meier curves for patients (n=205) with levels of NT-proBNP below and above the median value of 164 pg/ml (figure 1A) and the Kaplan-Meier curves for patients (n=205) with levels of PF4var/CXCL4Ll below and above the median value of 10 ng/ml (figure IB).

The results of the Cox proportional hazards models are shown in table 2. Models 1 and 2, including clinical factors (age, gender, creatinine, LVEF, medication) and either NT-proBNP or PF4var/CXCL4Ll, showed an additional prognostic value of either NT-proBNP or PF4var/CXCL4Ll. Model 3, including clinical factors and both NT-proBNP and PF4var/CXCL4Ll, showed additional prognostic value of PF4var/CXCL4Ll on top of NT-proBNP.

Figure 2 shows the Kaplan-Meier curves for patients with, respectively, NT-proBNP levels and PF4var/CXCL4Ll levels below and above the respective median values. The group of patients with NT-proBNP levels above the median value and PF4var/CXCL4Ll levels lower than the median value showed a significantly worse outcome as compared to the other groups (p<0.001).

The secondary outcome measure (MACE or hospitalization for congestive heart failure) was experienced by 40 patients. Figure 3 shows the Kaplan-Meier curves for patients with respectively NT-proBNP levels and PF4var/CXCL4Ll levels below and above the respective median values. Again, the group of patients with NT-proBNP levels above the median value and PF4var/CXCL4Ll levels lower than the median value showed a significantly worse outcome as compared to the other groups (p<0.001). Table 3 shows the event rates per patient year follow-up according to PF4var and NT- proBNP levels for the primary and secondary endpoint.

Conclusion

The chemokine platelet factor 4 variant (PF4var/CXCL4Ll) is a nonallelic variant of PF4/CXCL4 with lower affinity for heparin and higher angiostatic activity. The determinants and prognostic value of PF4var/CXCL4Ll in patients with heart disease, such as stable coronary artery disease (CAD) and preserved left ventricular (LV) function, were investigated. We evaluated 205 consecutive patients with stable CAD and preserved LV function (LV ejection fraction > 50%). Blood samples for PF4var/CXCL4Ll as well as NT-proBNP were taken at inclusion. Patients were followed (median follow-up 2.5 years) for the combined endpoint of cardiac death, non-fatal acute myocardial infarction, stroke or hospitalization for heart failure. Median PF4var/CXCL4Ll was 10 ng/ml (interquartile range 8-16 ng/ml). Independent determinants of PF4var/CXCL4Ll levels were age, gender and circulating platelet number. Patients who experienced cardiac events (n=20) during follow-up showed lower levels of PF4var/CXCL4Ll (8.5 [5.3-10] ng/ml versus 12 [8-16] ng/ml, p=0.033). ROC analysis for events showed an area under the curve (AUC) of 0.82 (95% CI 0.73-0.90, p<0.001) for higher NT-proBNP levels and an AUC of 0.32 (95% CI 0.19-0.45, p=0.009) for lower PF4var/CXCL4Ll levels. Cox proportional hazard analysis showed that PF4var/CXCL4Ll has an independent prognostic value on top of NT-proBNP. Low levels of PF4var/CXCL4Ll are associated with a poor outcome in patients with heart disease such as stable CAD and preserved LV function. This prognostic value is independent of NT-proBNP levels, suggesting that both neurohormonal and platelet related factors determine outcome in these patients.

Table 1: Clinical and laboratory characteristics of the total patients population and patients with PF4var/CXCL4Ll > 10 and < 10 ng/ml

Total Group PF4var/CXCL4Ll > PF4var/CXCL4Ll < p-value (n=205) 10 ng/ml 10 ng/ml

(n=107) (n=98)

Age 68±8 67±9 69±8 0.29

Gender (males) 172 (84%) 84 (79%) 89 (91%) 0.015

Previous AMI (%) 116 (57%) 60 (56%) 56 (57%) 0.89

Previous PCI (%) 71 (35%) 38 (36%) 33 (34%) 0.88

Previous CABG (%) 96 (47%) 47 (44%) 49 (50%) 0.40

Atrial fibrillation (%) 21 (10%) 10 (9%) 11 (11%) 0.82

Diabetes (%) 60 (30%) 35 (33%) 25 (26%) 0.28

Smoking (%) 22 (11%) 14 (13%) 8 (8%) 0.27

Hypertension (%) 131 (65%) 65 (62%) 66 (67%) 0.46

LVEF (%) 63±9 63±9 64±9 0.26

NYHA ll-ll l (%) 90 (44%) 45 (42%) 45 (46%) 0.74

6 minutes WD (m) 428±115 422±118 434±113 0.51

Creatinine (mg/dl) 1±0.22 0.97±0.20 1.06±0.24 0.004

NT-proBNP (pg/ml) 164 (79-354) 137 (69-291) 206 (100-415) 0.015 sTNFRI (pg/ml) 2.99±1.15 2.96±1.26 3.03±1.02 0.67 sTNFRII (pg/ml) 8.69±3.25 8.67±3.63 8.72±2.81 0.91

Medical treatment

Anti-platelets (%) 171 (84%) 90 (85%) 81 (82%) 0.94

Beta-blockers (%) 153 (75%) 75 (70%) 78 (80%) 0.15

ACE and/or ARB (%) 137 (67%) 73 (68%) 64 (65%) 0.41

Aldosterone antagonist (%) 14 (7%) 8 (8%) 6 (6%) 0.78

Coumarines (%) 36 (18%) 18 (17%) 18 (18%) 0.85

Statins (%) 133 (65%) 74 (69%) 59 (60%) 0.19

PF4var/CXCL4Ll (ng/ml) 10 (8-16) 15 (13-19) 7.5 (6-9) - PF4/CXCL4 (ng/ml) 3147 (2671-3772) 3260 (2919-3962) 2961 (2381-3493) 0.001

RANTES/CCL5 (pg/ml) 4044 (2405-6226) 4243 (2811-6687) 3662 (1913-6034) 0.028

Platelets x 10⁹ (/I) 228±57 239±58 215±54 0.003

AMI indicates acute myocardial infarction, PCI percutaneous coronary intervention, CABG coronary artery bypass grafting, LVEF left ventricular ejection fraction, NYHA New York Heart

Association, WD walking distance

P values indicate a statistical significant difference between patients groups with

PF4var/CXCL4Ll levels below and above the median value (10 ng/ml)

Table 2: Multivariate analysis: hazard ratios for NT-proBNP and PF4var/CXCL4Ll

Table 3. Event rates per patient year follow-up for the primary and secondary endpoint according to PF4var and NT-proBNP levels

Primary endpoint Secondary endpoint

(death due to cardiovascular causes - (death due to cardiovascular causes - non fatal AM I- non fatal stroke- non fatal AMI - PC - CABG - hospitalization congestive heart failure) hospitalization congestive heart failure)

N Cumulative Number Events/patie Cumulative Number Events/patie follow-up of events nt year FU follow-up of events nt year FU

(days) (days)

PF4var > median 64 79715 1 0.0045 74213 7 0.0344 and NT-proBN P <

median

42 44907 1 0.0081 41500 4 0.0352

PF4var < median

and NT-proBN P <

median 43 46348 3 0.0236 43115 6 0.0508 PF4var > median

and NT-proBN P > 56 51919 15 0.1054 45481 23 0.1846 median

PF4var < median

and NT-proBN P >

median

AM I indicates acute myocardial infarction, PCI percutaneous coronary intervention, CABG coronary artery bypass grafting, FU follow-up.

References

1 von Hundelshausen P, Petersen F, Brandt E. Platelet-derived chemokines in vascular biology.

Thromb Haemost 2007; 97:704-713.

2 Struyf S, Burdick MD, Proost P, Van Damme J, Strieter RM. Platelets release CXCL4L1, a nonallelic variant of the chemokine platelet factor-4/CXCL4 and potent inhibitor of angiogenesis. Circ Res

2004; 95:855-857.

3 Struyf S, Burdick MD, Peeters E et al. Platelet factor-4 variant chemokine CXCL4L1 inhibits melanoma and lung carcinoma growth and metastasis by preventing angiogenesis. Cancer Res 2007; 67:5940-5948.

4 Charo IF, Ransohoff RM. The many roles of chemokines and chemokine receptors in inflammation. N Engl J Med 2006; 354:610-621.

5 Coelho AL, Hogaboam CM, Kunkel SL. Chemokines provide the sustained inflammatory bridge between innate and acquired immunity. Cytokine Growth Factor Rev 2005; 16:553-560.

6 Bonecchi R, Galliera E, Borroni EM et al. Chemokines and chemokine receptors: an overview.

Front Biosci 2009; 14:540-551.

7 Lasagni L, Francalanci M, Annunziato F et al. An alternatively spliced variant of CXCR3 mediates the inhibition of endothelial cell growth induced by IP-10, Mig, and l-TAC, and acts as functional receptor for platelet factor 4. J Exp Med 2003; 197:1537-1549.

8 Mueller A, Meiser A, McDonagh EM et al. CXCL4-induced migration of activated T lymphocytes is mediated by the chemokine receptor CXCR3. J Leukoc Biol 2008; 83:875-882.

9 Petersen F, Bock L, Flad HD, Brandt E. A chondroitin sulfate proteoglycan on human neutrophils specifically binds platelet factor 4 and is involved in cell activation. J Immunol 1998; 161:4347- 4355.

10 Nesmelova IV, Sham Y, Gao J, Mayo KH. CXC and CC chemokines form mixed heterodimers: association free energies from molecular dynamics simulations and experimental correlations. J

Biol Chem 2008; 283:24155-24166.

11 Koenen RR, von Hundelshausen P, Nesmelova IV et al. Disrupting functional interactions between platelet chemokines inhibits atherosclerosis in hyperlipidemic mice. Nat Med 2009; 15:97-103.

12 Struyf S, Salogni L, Burdick MD et al. Angiostatic and chemotactic activities of the CXC chemokine CXCL4L1 (platelet factor-4 variant) are mediated by CXCR3. Blood 2011; 117:480-8.

13 Lazzeri E, Romagnani P. CXCR3-binding chemokines: novel multifunctional therapeutic targets.

Curr Drug Targets Immune Endocr Metabol Disord 2005; 5:109-118. Eslin DE, Zhang C, Samuels KJ et al. Transgenic mice studies demonstrate a role for platelet factor 4 in thrombosis: dissociation between anticoagulant and antithrombotic effect of heparin. Blood 2004; 104:3173-3180.

Greinacher A, Althaus K, Krauel K, Selleng S. Heparin-induced thrombocytopenia. Hamostaseologie 2010; 30:17-18.

Bibbins-Domingo K, Gupta R, Na B et al. N-terminal fragment of the prohormone brain-type natriuretic peptide (NT-proBNP), cardiovascular events, and mortality in patients with stable coronary heart disease. JAMA 2007; 297:169-176.

Van de Veire NR, De Winter O, Philippe J et al. Maximum oxygen uptake at peak exercise in elderly patients with coronary artery disease and preserved left ventricular function: the role of inflammation on top of tissue Doppler-derived systolic and diastolic function. Am Heart J 2006; 152:297.el-7.

Vandercappellen J, Noppen S, Verbeke H et al. Stimulation of angiostatic platelet factor-4 variant (CXCL4Ll/PF-4var) versus inhibition of angiogenic granulocyte chemotactic protein-2 (CXCL6/GCP-2) in normal and tumoral mesenchymal cells. J Leukoc Biol 2007; 82:1519-1530. Lasagni L, Grepin R, Mazzinghi B et al. PF-4/CXCL4 and CXCL4L1 exhibit distinct subcellular localization and a differentially regulated mechanism of secretion. Blood 2007; 109:4127-4134. Han ZC, Bellucci S, Tenza D, Caen JP. Negative regulation of human megakaryocytopoiesis by human platelet factor 4 and beta thromboglobulin: comparative analysis in bone marrow cultures from normal individuals and patients with essential thrombocythaemia and immune thrombocytopenic purpura. Br J Haematol 1990; 74:395-401.

Lambert MP, Sachais BS and Kowalska MA. Chemokines and thrombogenicity.Thromb Haemost. 2007; 97:722-9.

Sachais BS, Turrentine T, Dawicki McKenna JM, et al. Elimination of platelet factor 4 (PF4) from platelets reduces atherosclerosis in C57BI/6 and apoE-/- mice. Thromb Haemost. 2007; 98:1108- 13.

Scheuerer B, Ernst M, Durrbaum-Landmann I et al. The CXC-chemokine platelet factor 4 promotes monocyte survival and induces monocyte differentiation into macrophages. Blood 2000; 95:1158-1166.

Fricke I, Mitchell D, Petersen F et al. Platelet factor 4 in conjunction with IL-4 directs differentiation of human monocytes into specialized antigen-presenting cells. FASEB J 2004; 18:1588-1590. Gleissner CA, Shaked I, Little KM, Ley K. CXC chemokine ligand 4 induces a unique transcriptome in monocyte-derived macrophages. J Immunol 2010; 184:4810-4818.

Xia CQ, Kao KJ. Effect of CXC chemokine platelet factor 4 on differentiation and function of monocyte-derived dendritic cells. Int Immunol 2003; 15:1007-1015.

Woller G, Brandt E, Mittelstadt J, Rybakowski C, Petersen F. Platelet factor 4/CXCL4-stimulated human monocytes induce apoptosis in endothelial cells by the release of oxygen radicals. J Leukoc Biol 2008; 83:936-945.

Rudiger A, Fischler M, Harpes P et al. In critically ill patients, B-type natriuretic peptide (BNP) and N-terminal pro-BNP levels correlate with C-reactive protein values and leukocyte counts. Int J Cardiol 2008; 126:28-31.

Schnabel RB, Schulz A, Messow CM et al. Multiple marker approach to risk stratification in patients with stable coronary artery disease. Eur Heart J 2010.

Verbeke. Expression of angiostatic platelet factor-4var/CXCL4Ll counterbalances angiogenic impulses of vascular endothelial growth factor, interleukin-8/CXCL8, and stromal cell-derived factor 1/CXCL12 in esophageal and colorectal cancer. Hum Pathol 2010.

Claims

ims

1. In vitro use of the chemokine Platelet Factor 4 variant (PF4var) comprising the amino acid sequence as depicted by SEQ ID N°l or a fragment thereof comprising at least the amino acid sequence as depicted by SEQ ID N°2 as a marker for the prognosis of cardiovascular outcome in a patient having heart disease.

2. In vitro use according to claim 1, wherein said heart disease is coronary artery disease with preserved left ventricular function.

3. In vitro use according to claim 1 or 2, wherein said chemokine is used together with another marker for the prognosis of cardiovascular outcome in patients having heart disease.

4. In vitro use according to claim 3, wherein said other marker is N-terminal proBrain natriuretic peptide (NT-proBNP).

5. An in vitro method to prognose the cardiovascular outcome in a patient having heart disease comprising the steps:

1) obtaining a sample from said patients,

2) determining the level of PF4var having the amino acid sequence as depicted by SEQ ID N°l or a fragment thereof comprising at least the amino acid sequence as depicted by SEQ ID N°2,

and 3) determining if the level of PF4var leads to a negative prognosis for said patient if the level determined in step 2) is lower than the median PF4var level of the population to which said patient belongs.

6. The in vitro method according to claim 5 , wherein said sample is a blood sample.

7. The in vitro method according to any of claims 5 to 6, wherein said level is determined via a sandwich-type ELISA.

8. The in vitro method according to any of claims 5 to 6, wherein said level is determined via a quantitative T-PC assay.

9. Use of a kit in the prognosis of the cardiovascular outcome in a patient having heart disease wherein said kit comprises reagents to perform an in vitro assay for determining the level of PF4var having the amino acid sequence as depicted by SEQ ID N°l, or a fragment thereof comprising at least the amino acid sequence as depicted by SEQ ID N°2, in a sample obtained from said patient.

10. Use of a kit according to claim 9, wherein said assay is a sandwich-type ELISA or a quantitative RT-PCR assay. Use of a kit according to claim 9 or 10, wherein said reagents at least comprise an antibody which specifically detects PF4var having the amino acid sequence as depicted by SEQ ID N°l, or a fragment thereof comprising at least the amino acid sequence as depicted by SEQ ID N°2; or a nucleic acid probe which specifically detects the PF4var nucleic acid such as the PF4var m NA.