WO2011135275A1 - Method and microbubbles for detecting atherosclerotic plaque - Google Patents

Abstract

The invention relates generally to methods of diagnosing an increased risk of a neurovascular event such as a stroke in a subject. The methods comprise administering ultrasound contrast microbubbles to the subject, and, if an atherosclerotic plaque is present in the subject, using late phase contrast enhanced ultrasound to detect a signal produced by microbubbles in the plaque, wherein the higher the signal, the greater the risk of that subject suffering a neurovascular event such as a stroke.

Description

METHOD AND MICROBUBBLES FOR DETECTING ATHEROSCLEROTIC PLAQUE

The present invention relates generally to methods for diagnosing a subject's risk of suffering a neurovascular event such as stroke, by assessing atherosclerotic plaques present in that subject. Preferably, the methods of the invention involve the detection of microbubbles retained within atherosclerotic plaques, using ultrasound techniques such as Late Phase Contrast Enhanced Ultrasound (LP-CEUS).

The occurrence of damaging neurovascular events such as stroke is becoming increasingly prevalent, particularly in Western societies. Large artery atherosclerosis accounts for approximately 30% of strokes (Petty et al,. Stroke 1999

Dec;30(12):2513-6). The mechanism by which stroke occurs is normally rupture of carotid atherosclerotic plaques, when the integrity of the fibrous cap of the plaque is lost and luminal blood communicates with the plaque's thrombogenic core (Virmani, et al, J Am Coll Cardiol 2006 Apr 18;47(8 Suppl):C13-C18). This leads to the formation of a thrombus, which can subsequently embolise and occlude a distal cerebral artery, resulting in stroke.

It has been suggested that plaques most vulnerable to rupture are those which protrude to a greater degree into the vascular lumen. Accordingly, current clinical

investigations aiming to assess the risk of atherosclerotic plaque rupture use imaging studies to quantify the extent to which the plaques protrude into the lumen (luminal stenosis). Such imaging techniques include fluorodeoxyglucose positron emission tomography (Rudd et al, Circulation 2002 Jun 11;105(23):2708-11) and magnetic resonance imaging (MRI) enhanced with either iron oxide particles (Kooi et al, Circulation 2003 May 20;107(19):2453-8.; Trivedi et al, Circulation 2003 Nov 1 l;108(19):el40) or gadolinium (Kerwin et al, Radiology 2006 Nov;241(2):459-68.). However, the high cost of these techniques prohibits them from being translated into routine clinical practice to assess a subject's risk of stroke or other neurovascular event. In addition, there is evidence to suggest that atherosclerotic plaques most at risk of rupture are not necessarily those which impinge most substantially upon the lumen (Topol & Nissen, Circulation 1995 Oct 15;92(8):2333-42). Furthermore, the currently used imaging methods are relatively poor at predicting which patients who have not previously suffered any neurovascular event are likely to experience a neurovascular event such as stroke, since these methods do not provide any functional information about the atherosclerotic plaques. Therefore, there is a need for alternative methods of assessing the potential occurrence of a neurovascular event such as a stroke, which do not rely solely on a visual assessment of atherosclerotic plaques.

In a first aspect, the invention provides a method of diagnosing an increased risk of a neurovascular event in a subject, the method comprising:

i) administering microbubbles to the subject;

ii) if an atherosclerotic plaque is present in the subject, using late phase contrast enhanced ultrasound to detect a signal produced by microbubbles in the plaque;

wherein the higher the signal, the greater the risk of that subject suffering a neurovascular event.

The present invention is based on the surprising finding that ultrasound contrast agents known as microbubbles are retained within atherosclerotic plaques. Thus, the method of the invention involves the detection of ultrasound contrast microbubbles retained within an atherosclerotic plaque. The inventors have found that a higher signal produced by microbubbles retained within an atherosclerotic plaque correlates with an increased risk of suffering a neurovascular event. Interestingly, the microbubble signal may not directly reflect the extent to which the plaque impinges upon the lumen. The detection of retained microbubbles is therefore particularly useful in the diagnosis of a subject's risk of suffering a neurovascular event such as stroke. Microbubbles are ultrasound contrast agents comprising an outer shell and a gas core, which show a high degree of echogenicity and are therefore suitable contrast agents for use in ultrasound scans.

Microbubbles have been used to detect the rate of vascular blood flow, for example in sclerotic arteries. In addition, microbubbles have been used as a blood pool contrast agent to detect the rate of blood flow through atherosclerotic plaques (Xiong et al, Radiology 2009 May;251 (2):583-9). It is known that microbubbles can be phagocytosed by monocytes in vitro and can remain acoustically active for up to 30 minutes (Lindner et al ., Circulation 2000 Aug 1 ;102(5):531-8; Yanagisawa et al, Ultrasound Med Biol 2007 Feb;33(2):318-25). Previous studies have shown that microbubbles can be detected on leukocytes which are attached to the endothelium of inflamed tissue (Lindner et al, Circulation 2000 Feb 15;101(6):668-75). It has also been demonstrated that microbubbles attach directly to activated leukocytes adherent to the surface of damaged endothelium in mice (Tsutsui et al, J Am Coll Cardiol 2004 Sep 1 ;44(5): 1036-46). However, microbubbles have never previously been shown to be retained within atherosclerotic plaques themselves (as opposed to simply adhering to the endothelial surface or to activated leukocytes attached to the endothelial surface) and certainly not in sufficiently detectable amounts to suggest that they could be used in clinical applications to assess the extent of plaque formation.

Any known microbubbles can be used in the present invention. For example, commercially available microbubbles such as SonoVue™ can be used. The microbubble shells may comprise albumin, galactose, one or more lipids or one or more polymers. Preferably, the microbubble shells comprise one or more lipids.

Microbubble shells typically contain air or a heavy gas, which oscillates when subjected to an ultrasonic frequency, thereby producing a characteristic echo which can be detected during an ultrasound scan. Suitable gases include perfluorocarbon, nitrogen, sulphur hexafluoride and octafluoropropane (perflutren), amongst others. Preferably, the microbubbles contain sulphur hexafluoride gas.

The microbubbles used in the present invention can be untargetted or targetted.

Targetted microbubbles contain exposed surface ligands that bind specifically to particular receptors, or exposed surface receptors that bind specifically to particular ligands. Untargetted microbubbles contain no such ligands/receptors. Surprisingly, the inventors have shown that even untargetted microbubbles are retained in atherosclerotic plaques in sufficient numbers to allow their detection by ultrasound. Thus, microbubbles which are specifically adapted to target plaque components are likely to further improve the sensitivity of the methods of the invention.

Targetted microbubbles which may be particularly useful in the present invention may comprise ligands such as substrates, monoclonal antibodies, activators, inhibitors, neurotransmitters, or receptors such as selectins, vascular endothelial growth factor (VEGF), vascular cell adhesion molecule 1 (VCAM-1), inter-cellular adhesion molecule 1 (ICAM-1), and others.

Microbubbles are preferably administered intravenously into a subject. Preferably, the microbubbles are administered in the form of a bolus injection. The microbubbles may be prepared for injection by suitable means known in the art. For example, the microbubbles may be prepared by reconstituting a lyophilised microbubble preparation in a suitable solvent. Once present in a subject's circulatory system, the microbubbles generally have a half life ranging from about 1.5 to about 3 minutes. Thus, previous applications of microbubbles in measuring blood flow have required that an ultrasound scan is performed within about 3 minutes of administration of the microbubbles to the patient. It was previously believed that the short half life and rapid dissipation of microbubbles within a subject's circulatory system meant that any analysis of blood flow had to be performed almost immediately after the

microbubbles had been administered. The inventors have shown for the first time that microbubbles can be retained within atherosclerotic plaques themselves. Whilst not wishing to be limited by theory, this may occur by direct perfusion of the microbubbles into the plaque core, by phagocytosis of the microbubbles by leukocytes which subsequently invade the plaque, or by other means. The microbubbles may accumulate predominatly in the upstream half of the plaque where the inflammation is maximal. Accordingly, the methods of the invention may involve measuring a signal produced by microbubbles present in an upstream portion (such as the upstream half) of an atherosclerotic plaque.

The signal produced by microbubbles retained in an atherosclerotic plaque can be detected after the majority of circulating bubbles have expired, or have left the particular region of the circulatory system comprising the plaque. LP-CEUS is therefore particularly suited to detecting the level of retained microbubbles.

The late phase contrast enhanced ultrasound performed in the methods of the present invention may involve the application of an ultrasound scanner to a region of the subject's anatomy suspected of containing an atherosclerotic plaque at any time after about 4, 5, 6, 7, 8, 9 or 10 minutes of administration of the microbubbles. Preferably, the ultrasound scan is performed between about 6 and about 10 minutes after administration of the microbubbles. The ultrasound scan may be performed between about 4.5 and about 7.5 minutes, or between about 5 and about 7 minutes, or between about 5.5 and about 6.5 minutes after administration of the microbubbles. Most preferably, the ultrasound scan is performed about 6 minutes after administration of the microbubbles.

Suitable ultrasound scanners for use in the present invention are commercially available. Preferably, the ultrasound scanner has a high frequency linear array LI 2-5 or L9-4 MHz probe.

Preferably, LP-CEUS is performed using flash-imaging at an intermediate mechanical index._' For example, the mechanical index used can range from about 0.15 to about 0.45. Preferably, a mechanical index of about 0.34 is used. An intermediate mechanical index is advantageous in that it minimises artefacts produced during the flash-imaging process. Preferably, a non-linear imaging contrast mode is used. Alternatively, phase inversion or power modulation or combined pulse inversion and power modulation can be used. An advantage of using a non-linear contrast mode is that this can result in the complete subtraction of signal produced by the native tissue. The ultrasound scan produces an image from which the microbubble signal can be quantified. For example, the echo intensity of the microbubbles can be quantified by measuring the contrast of the image. A single two-dimensional or three-dimensional image can be used to quantify the signal. Alternatively, a number of two-dimensional or three-dimensional images within a specified time frame can be captured and the signal can be determined by analysis of each of those images. For example, an ultrasound scanner can be used to capture a series of images taken at about 0.01 to about 1 second intervals of the plaque and adjacent lumen. Preferably, about six images are captured within about one second. Since the timeframe required for image acquisition in LP-CEUS is shorter than the timeframe required for dynamic contrast enhanced ultrasound imaging methods that are used to assess blood flow, the examination is technically easier to perform. This renders the methods of the present invention particularly suitable for clinical applications, since medical practitioners do not require extensive training in order to perform the examination. In addition, the shorter examination time improves patient comfort during examination. Furthermore, since image acquisition during LP-CEUS can last a relatively short time (such as less than 1 second), motion artefacts do not cause a problem in the images produced. The ultrasound images can be recorded electronically on any computer-readable medium and stored for subsequent analysis. Where a series of images is taken, this series can be successively displayed as a video loop. The signal produced by the microbubbles can be quantified by a number of means. Preferably, computer software programs such as the QLAB software (Philips; Bothel, WA) can be used to anaylse echo intensity of a particular region of the image.

Alternative software packages are available. In order to quantify the microbubble signal, a medical practitioner can mark an area of the image representing the atherosclerotic plaque and the average intensity produced in that area can be determined (the "plaque signal"). The practitioner can also mark an area of the image representing the lumen of the vasculature adjacent the plaque and the average intensity produced in that area can also be determined (the "lumen signal"). These two average values can be used to produce a normalised value of microbubble signal by dividing the plaque signal by the lumen signal, thereby accounting for any background" signal produced in the lumen. This calculation provides a single statistical value quantifying the signal produced by microbubbles present in the plaque, which can be used to determine the subject's risk of suffering a neurovascular event.

The simple production of a single statistical value from which a subject's risk of suffering a neurovascular event can be determined allows a practitioner to carry out the invention without needing to perform detailed calculations and without requiring extensive statistical training.

Of course, the single statistical value does not need to be the normalised value produced by dividing the average plaque signal by the average lumen signal. Other statistical values can be used. For example, the mean plaque signal itself can be used. This value may be sufficient to indicate a subject's risk of suffering a neurovascular event. Alternatively, other calculations can be performed on the data produced by the ultrasound scan to produce a single value indicative of the subject's risk of suffering a neurovascular event. Statistical software packages may be used to prepare such a value. Gray-scale median (GSM) can be used to quantify the signal produced by the plaque constituents, for example, as described in Sabetai MM, et al. (Stroke Sep;31(9):2189- 96 (2000)), as an alternative or additional measurement to measuring echo intensity. A medical professional can use the microbubble signal produced from a subject's atherosclerotic plaque to determine the risk of that subject suffering a neurovascular event in the future. This risk assessment may involve comparing the subject's microbubble signal with a calibrated data set indicating the probability of a given subject suffering a neurovascular event. For example, a calibration curve can be produced correlating microbubble signal with risk of suffering a neurovascular event. Such a calibrated data set can be produced, for example, by performing the method of the invention on a first group of subjects known to have an atherosclerotic plaque and known to have previously suffered a neurovascular event. In these subjects, the risk of suffering a repeated neurovascular event is known to be increased. A further group of subjects can be chosen who have atherosclerotic plaques but have never previously suffered a neurovascular event. In this group of subjects, the risk of suffering a neurovascular event is lower than that in the first group of subjects. The method of the invention can be performed on each group of subjects and a range of microbubble signals recorded for each group.

When the method of the invention is subsequently performed on a subject with an atherosclerotic plaque, the risk of that subject suffering a neurovascular event in future can be determined by comparing the microbubble signal value in that subject with the results obtained from the two "calibration" groups. If the subject's microbubble signal falls within the range of the first calibration group, the risk of that subject suffering a neurovascular event may be considered to be high, potentially warranting aggressive medical treatment to reduce that risk immediately.

Alternatively, if a subject's microbubble signal lies above a threshold value on a calibration curve correlating microbubble signal with risk of suffering a neurovascular event, an appropriate course of therapy can be selected accordingly. The threshold value can be determined by a medical professional, and the specific value may vary depending on the particular experimental protocol followed in measuring microbubble signal.

A more accurate calibration may be achieved by performing the method of the invention on subjects that have suffered specific neurovascular events such as stroke, which may be categorised according to severity of the stroke. The location of the plaque in subjects who have previously suffered a stroke may also be taken into account when performing the calibration. In addition, the time between a subject suffering a neurovascular event and being evaluated in the present invention in order to produce a calibrated data set can be taken into account, thereby potentially giving a more accurate evaluation of the relationship between microbubble signal and risk of suffering a neurovascular event.

These additional steps of preparing a calibrated data set against which an individual subject's microbubble signal can be compared may be included in the method of the invention. Thus, the invention may further comprise preparing a calibrated data set relating microbubble signal to risk of suffering a neurovascular event.

Upon obtaining a microbubble signal from a subject and assessing the risk of that subject suffering from a neurovascular event, a medical professional can decide a suitable course of treatment for that subject. If the risk is high, the medical professional can advise a course of aggressive treatment (such as surgery,

endovascular stent placement, or other treatments) to immediately reduce the risk of that subject suffering a neurovascular event. Thus, a medical professional can use the method of the present invention to select an appropriate therapy for a subject.

The present invention can be used to diagnose the risk of a subject suffering neurovascular events such as stroke, transient ischaemic attack (TIA), amaurosis fugax, thrombosis, infarction, or others.

The present invention also provides microbubbles for use in diagnosing an increased risk of a neurovascular event in a subject. The microbubbles can be provided for use in any of the methods described herein. Any of the microbubbles as described herein can be used.

Furthermore, the present invention provides the use of microbubbles in the manufacture of a diagnostic for diagnosing an increased risk of a neurovascular event in a subject. The diagnostic can be provided for use in any of the methods described herein. Any of the microbubbles as described herein can be used.

Preferred features of each aspect of the invention are as for each of the other aspects mutatis mutandis. The prior art documents mentioned herein are incorporated to the fullest extent permitted by law.

The invention is further illustrated by the following non-limiting examples and by reference to the Figures in which:

Figure 1 shows conventional ultrasound of the plaque (grey arrow head) and carotid lumen (white arrow head).

Figure 2 shows the corresponding plaque as shown in Figure 1 with microbubbles present in the plaque (grey arrow head) and in the carotid lumen (white arrow head).

Figure 3 shows linear regression analysis relating levels of specific cytokines and chemokines to atherosclerotic plaque LP-CEUS normalised signal. Example 1

The following study was performed in order to determine whether non-targeted microbubbles are retained in human carotid plaque in sufficient number to be detected by ultrasound.

Study Subjects: This prospective study was approved by the local Research Ethics Committee and the Medicines and Healthcare Products Regulatory Agency (MHRA). MHRA approval was required because the contrast agent was being used for an off label indication. Informed consent was obtained from all subjects prior to their examination. Between December 2008 and May 2009, subjects aged 18 years or over who presented to the vascular clinic for a carotid duplex ultrasound scan at our institution, and whose scan showed an atherosclerotic plaque of > 30% stenosis by velocity criteria were recruited. Subjects with myocardial infarction or unstable angina within 14 days, cardiac failure (NYHA Class III/IV) or prosthetic heart valves were excluded as these are contraindications to receiving the contrast agent. 37 subjects, with a mean age of 69.9 years (+/- 8.5 years standard deviation) were enrolled. 27 (73%) were male, with a mean age of 69.7 years (range 58-86) and 10 (27%) were female, with a mean age of 70.3 years (range 49-86) with no significant difference in the ages of men and women).

Plaques were defined as "symptomatic" if symptoms consistent with stroke, transient ischaemic attack (TIA) or amaurosis fugax had occurred within 12 months before entry into the study, in the neurovascular territory of the plaque studied. Plaques were defined as "asymptomatic" if no such events had ever occurred within the neurovascular territory of the plaque studied. Of the 37 subjects, 16 subjects were recruited into the symptomatic group and 21 into the asymptomatic group (Table 1).

Table 1: Characteristics of Symptomatic and Asymptomatic Groups

Note - unless otherwise indicated, data are numbers of patients, with percentages in parentheses.

Of the symptomatic group, the time from cardiovascular event to LP-CEUS assessment was less than 30 days for 10 subjects (63%), and less than 50 days for 14 subjects (88%). With regard to the cerebrovascular events, 6 subjects (37.5%) had had a stroke, 8 (50%) had had a TIA, and 2 (12.5%) had had amaurosis fugax. Of the asymptomatic group, 7 (33%) had previously been diagnosed with TIA or stroke which affected the hemisphere contralateral to the plaque which was studied. The remaining 14 (67%) of the asymptomatic group had never been diagnosed with TIA, stroke or amaurosis fugax. There was no significant difference between the symptomatic and asymptomatic groups iri age, gender, or history of diabetes mellitus, hypertension, presence of an HMG-CoA reductase inhibitor (statin) on their prescription, and smoking (Table 1).

Standard and Contrast-enhanced carotid ultrasound:

The ultrasound examination was performed with the subject in the supine position using a Philips (Bothel, WA) iU22 ultrasound scanner equipped with a high frequency linear array LI 2-5 MHz probe. Luminal stenosis was measured in the sagittal plane using the velocity criteria which approximates the NASCET criteria (Oates et al, Eur J Vase Endovasc Surg 2009 Mar;37(3):251-61; Sidhu et al, Clin Radiol 1997 Sep;52(9):654-8). This measurement was made by a clinical vascular scientist (with a minimum of 5 years experience in carotid ultrasound) as part of the subjects' routine care. The Gray-scale median score was calculated using the luminal blood and carotid wall adventia as reference points for normalisation (Sabetai et al, Stroke 2000 Sep;31(9):2189-96).

LP-CEUS was subsequently performed. The contrast agent used was SonoVue™ (Bracco spa, Milan, Italy) which consists of a phospholipid shell containing the inert gas sulphur hexafluoride. The agent is prepared immediately prior to the examination by mixing 25mg of the lyophilisate powder with 5mL of saline. 2mL of this preparation was injected as an intravenous bolus into an antecubital vein. Subjects were observed for 30 minutes following administration of the contrast agent and verbally asked about the occurrence of adverse events.

LP-CEUS was performed with flash-imaging at intermediate mechanical index (MI: 0.34) of the carotid bifurcation and internal carotid artery, using a non-linear imaging (power modulation) contrast mode, 6 minutes following the bolus contrast injection. Six flash frames were acquired in less than 1 second in the axial orientation at the level of greatest stenosis. The cine-loop of the acquisition was saved on the hard- drive. Contrast enhanced scans were performed by a radiologist with 3 years of carotid ultrasound experience who was blinded to the subjects' clinical information.

QLAB software (Philips; Bothel, WA) was used to quantify echo intensity of the plaque from the acquired cine-loop and to measure Gray-scale median. Raw linear data was used for analysis. All regions of interest (ROIs) were drawn by an ultrasonologist with 17 years experience who was blinded to medical history. Using the fundamental B mode image, a single ROI was drawn on the outline of the plaque (Figures 1 and 2). The ROI is automatically mapped to the same position on the contrast image, and QLAB calculates the signal intensity of each pixel within the ROI. The mean signal intensity is then automatically calculated. A second ROI was drawn around the residual lumen to calculate mean signal intensity in the lumen.

Statistical Analysis:

The raw linear data generated from QLAB was found to be log-normally distributed, and therefore was log transformed for statistical analysis. The signal intensity of the plaque was normalised by dividing the plaque signal intensity by the lumen signal intensity (because the data was log transformed, normalisation required subtracting the lumen signal intensity from the plaque signal intensity). The normalised signal was compared between the two groups (symptomatic and asymptomatic) using the t- test, assuming unequal variances (GraphPad Prism 5.01, GraphPad Software, San Diego, CA). Within the symptomatic group, plaques in the territory of a stroke were compared to those in the territory of a TIA. Within the asymptomatic group, plaques from subjects who had never had a cerebrovascular event were compared to those who had a history of such an event affecting the contralateral hemisphere. These comparisons were made using the t-test. Pearson correlation was used to investigate both the relationship between gray scale median and LP-CEUS signal, and the relationship between LP-CEUS signal and luminal stenosis. Baseline frequencies between the two groups were compared using Fisher's exact test or t-tests where appropriate. The possible relationship between subject characteristics and LP-CEUS was assessed using analysis of variance (ANOVA) with the characteristic as a covariate (SAS 9.1.3, SAS Institute Inc., Cary, NC). Sensitivity and Specificity of LP- CEUS for correctly identifying plaques as symptomatic or asymptomatic were derived by using receiver operating characteristic (ROC) curve analysis. Cut-off values were chosen which minimised the difference between sensitivity and specificity. A p value of 0.05 was used to determine significance for all statistical tests. The number of subjects enrolled in the study provided 90% power to detect effect sizes

(difference/standard deviation) of 1.1 with a 5% type-I error rate. Sample size calculations were performed using PASS (NCSS, Kaysville, Utah).

Results:

Of the 37 subjects, 16 subjects (43%) had plaques which were defined as symptomatic and 21 (57%) had plaques which were defined as asymptomatic. All scans produced evaluable LP-CEUS data. Normalised late phase plaque intensity was greater in the symptomatic group 0.3899 (95% CI: -0.1056 to 0.8854) than the asymptomatic group 0.6869 (95% CI: -1.036 to -0.3380), (P=0.0005). There was a moderate (rho=-0.44, P=0.016) inverse correlation between normalised late phase plaque intensity and gray scale median score.

Relationship between LP-CEUS signal and Symptoms:

Of the 37 subjects enrolled, all produced evaluable LP-CEUS data and none reported an adverse event during the observation period. Percentage luminal stenosis ranged from 30-99%, and whilst the mean stenosis in the symptomatic group was greater than in the asymptomatic group, the difference between the two groups did not achieve statistical significance (P=0.06) (Table 2). Table 2: Ultrasound Features of Carotid Plaque in Patients with and without

Symptoms

Data is expressed as: Mean (95% confidence interval)

Given the standard deviation observed in this study and the sample size, a difference of 20.5% in stenosis would have been required to provide 90% power to detect a difference between the two groups. The LP-CEUS normalised plaque intensity was significantly greater in the symptomatic group 0.3899 (95% CI: -0.1056 to 0.8854) than the asymptomatic group -0.6869 (95% CI: -1.036 to -0.3380), (P=0.0005). However, there was overlap in signal intensity between the two groups. Of note, the lowest signal intensity from the symptomatic group was derived from the subject who had the longest event to scan time (of nearly 1 year). For this subject, the signal fell below the mean of the asymptomatic group.

Relationship between LP-CEUS and sonographic findings:

There was a significant difference between the two groups with regard to gray scale median. The asymptomatic group had a higher gray scale median score than the symptomatic group (asymptomatic mean 29.01 (95% CI: 32.844 to 25.176), symptomatic mean 16.95 (95% CI: 18.544 to 15.356) (P=0.0124)). There was a moderate (rho^-0.44, P=0.016) inverse correlation between normalised plaque intensity and gray scale median score, demonstrating the tendency for plaques with greater normalised plaque intensity to have a lower gray scale median score. There was no correlation between normalised LP-CEUS plaque intensity and percentage luminal stenosis (P=0.27).

Relationship between LP-CEUS and Subject Characteristics:

There was no evidence of a relationship between diabetes, smoking history, or statin dose with LP-CEUS signal, nor of differential LP-CEUS signal between those subjects in the symptomatic group with TIA versus CVA. Within the asymptomatic group, there was no significant difference in LP-CEUS signal between those subjects who had never had a cerebrovascular event and those who had a history of such an event affecting the contralateral hemisphere.

Receiver operating characteristic curve analysis:

Sensitivity and specificity were 75% and 86% respectively for a LP-CEUS normalised peak intensity cut off > 0.

Conclusion:

By quantifying microbubble retention within the carotid plaque, LP-CEUS is able to show clear differences between groups of subjects with plaque ipsilateral to symptoms and asymptomatic plaques. This technique is therefore useful as a tissue specific marker of inflammation, and has a potentially important role in the risk stratification of atherosclerotic carotid stenosis. This study has demonstrated that the LP-CEUS signal intensity of carotid plaques is greater from plaques which are symptomatic (i.e. within the neurovascular territory of a recent cerebrovascular event), compared to those which are asymptomatic. This suggests that plaques responsible for cerebrovascular events are those that tend to have late sonographic enhancement.

It is unlikely that this finding merely reflects the possibility that symptomatic plaques have more intraplaque blood volume than asymptomatic plaques, and thus contain more circulating microbubbles in the late phase because, during the dynamic phase, the signal from the lumen is approximately two orders of magnitude greater than the signal from the plaque. Were the late phase plaque signal purely a result of circulating microbubbles, it should only represent a small fraction of the late phase lumen signal. The fact that the late phase signals of plaque and lumen are of similar magnitude demonstrates that microbubbles have accumulated within the plaque. We therefore suggest that by detecting retained untargeted microbubbles, LP-CEUS is able to detect inflammation and/or endothelial activation within carotid plaques in vivo. This is believed to be the first study demonstrating microbubbles passively targeting atherosclerotic carotid plaque in humans. This finding may have important clinical consequences for patients at high risk of plaque rupture and consequent stroke, because ex vivo analysis of atherosclerotic plaques has demonstrated that plaques causing rupture are characterised by an abundance of macrophages and an

inflammatory infiltrate (Virmani et al, J Am Coll Cardiol 2006 Apr 18;47(8

Suppl):C13-C18;4-7; Schaar et al, Eur Heart J 2004 Jun;25(12):1077-82; van der Wal et al., Circulation 1994 Jan;89(l):36-44; Boyle J Pathol 1997 Jan;181(l):93-9; Moreno et al ., Circulation 1994 Aug;90(2):775-8). The fact that the signal intensity derived from the symptomatic group overlaps with the asymptomatic group shows that LP-CEUS has the potential to identify asymptomatic patients who are at high risk of cerebrovascular events who might benefit from intensive medical treatment or surgical intervention.

Gray scale median score was greater in asymptomatic compared to symptomatic plaques. Although the strength of the correlation was only moderate, the LP-CEUS signal was inversely associated with gray scale median score. We believe that plaques with a low gray scale median score may have high lipid, haemorrhage and

macrophage content and may therefore retain more microbubbles. In conclusion, by quantifying microbubble retention within the carotid plaque, LP- CEUS is able to show clear differences between groups of plaques within the neurovascular territory of recent cerebrovascular events and asymptomatic plaques; it thus has promise as a tissue specific marker of inflammation. This technique may be useful in helping to identify those asymptomatic patients who might benefit from intensive medical or surgical therapy, or as a biomarker to investigate

pharmacodynamic effects of experimental molecules.

Example 2

The following multi-analyte profiling study was performed in order to confirm that LP-CEUS detection of microbubbles is an accurate indicator of inflammation. Carotid endarterectomy was performed on those candidates who had been subjected to LP-CEUS as described in Example 1. From the carotid atherosclerotic plaques, cells were isolated as a mixed cell suspension and cultured at lxlO⁶ cells/mL for 24 hours, at which point the supernatant was aspirated. In this unstimulated system, the cells displayed a spontaneous cytokine and chemokine production. There was a significant relationship between LP-CEUS normalised signal and the levels of IL6, IL 10, GM-CSK, IP 10 and RANTES, indicating that LP-CEUS signal accurately corresponds with levels of pro-inflammatory markers.

Results:

The linear regression analysis relating levels of the following cytokines and chemokines to atherosclerotic plaque LP-CEUS normalised signal shown in Figure 3 is summarised in Table 3.

Table 3: Linear regression analysis comparing analvte levels with LP-CEUS result.

Inflammation marker Regression analysis result

IL6 n=13; r2= 0.2645, p= 0.0072

IL10 n=12, r2= 0.3982, p=0.0009

GM-CSF n=13, r2=0.1729, p=0.0387

IP10 (CXCL10) n=8, r2=0.4973, p=0.0071 RANTES (CCL5) n=11 , r2=0.2, p=0.039

Thus, each of the pro-inflammatory markers above significantly correlated to the LP- CEUS results, indicating that LP-CEUS is an accurate indicator of plaque

inflammation.

In view of the fact that IL10 is generally considered an anti-inflammatory molecule, although has been shown to be elevated in areas where inflammatory disregulation is in existence, ILlO/TNFa ratio was calculated and confirmed a significant negative association with LP-CEUS signal. Linear regression: n=12, r2=0.2827, p=0.0075; hence a net pro-inflammatory situation.

The invention has been described in language more or less specific as to the structural features of the invention. It is to be understood, however, that the invention is not limited to the specific features described, since the means and construction shown comprise preferred forms of putting the invention into effect. The invention is, therefore, claimed in any of its forms or modifications within the legitimate and valid scope of the appended claims, appropriately interpreted in accordance with the doctrine of equivalents.

Claims

Claims:

1. A method of diagnosing an increased risk of a neurovascular event in a subject, the method comprising:

i) administering microbubbles to the subject;

ii) if an atherosclerotic plaque is present in the subject, using late phase contrast enhanced ultrasound to detect a signal produced by microbubbles in the plaque;

wherein the higher the signal, the greater the risk of that subject suffering a neurovascular event.

2. The method of claim 1 , wherein step i) comprises injecting microbubbles into the subject intravenously.

3. The method of any preceding claim, wherein step ii) comprises performing an ultrasound scan of the atherosclerotic plaque from about six to about ten minutes after administration of the microbubbles.

4. The method of claim 3, wherein the ultrasound scan is performed about six minutes after administration of the microbubbles.

5. The method of any preceding claim, wherein the signal is detected in step ii) by quantifying the echo intensity of the plaque from an ultrasound image of the plaque.

6. The method of any preceding claim, wherein the signal is calculated using computer software.

7. The method of any preceding claim, wherein a statistical value representing the signal is produced by dividing the mean echo intensity of the plaque by the mean echo intensity of the lumen of a blood vessel comprising the plaque.

8. The method of any preceding claim, wherein the microbubbles are phospholipid shells containing sulphur hexafluoride gas.

9. The method of any preceding claim, wherein the microbubbles are SonoVue™ microbubbles.

10. The method of any preceding claim, wherein the neurovascular event is a stroke.

11. The method of any preceding claim, wherein step i) is performed with an ultrasound scanner.

12. The method of claim 11 , wherein the ultrasound scanner has a high frequency linear array LI 2-5 MHz probe.

13. Microbubbles for use in diagnosing an increased risk of a neurovascular event in a subject.

14. The microbubbles of claim 13, for use in the method of any one of claims 1- 12.