PET imaging of aortic atherosclerosis: Is combined imaging of plaque anatomy and function an amaranthine quest or conceivable reality?
- 245 Downloads
Traditionally, blood vessels have been studied using contrast luminography to determine the site, extent and severity of luminal compromise by atherosclerotic deposits. Similar anatomical data can now be acquired non-invasively using ultrasound, computed tomography or magnetic resonance imaging. Plaque stability is an important determinant of subsequent vascular events and currently functional data on the stability of plaque is less well provided by these methods. The search for non-invasive techniques to image combined plaque anatomy and function has been pursued with visionary anticipation. This expectation may soon be realised as imaging with radionuclide-labelled atheroma-targeted contrast agents has demonstrated that plaque functional characteristics can now be shown. Increasingly positron emission tomography/computed tomography (PET/CT) imaging with 18F fluorodexoyglucose (FDG) and other radionuclides is being used to determine culprit plaques in complex clinically scenarios. Clinically, this information may prove extremely valuable in the assessment of stable and unstable patients and its use in prime time medical practice is eagerly awaited. We will discuss the current clinical applications of functional atheroma imaging in the aorta and highlight the promising preclinical data on novel image biomarkers of plaque instability. If clinical science is able to successfully translate these advances in vascular imaging from the bench to the bedside, a new paradigm will be achieved in cardiovascular diagnostics.
KeywordsAorta atherosclerosis functional molecular imaging PET
A perplexing quest in vascular imaging is the desire to identify plaque anatomy and function non-invasively. Such images have been sought by cardiologists and imaging physicians alike in a search for pictures that would be valuable in clinical diagnosis and management. “Amaranthine” describes the quest well for it derives from the Greek for unfading (amarantos) and flower (anthos) and is used to describe a genus of plant recognised for its unfading flowers. In Greek mythology amaranthine came to symbolise immortality. This particular search in vascular imaging has been amaranthine in its length, while beautiful in its promise and is becoming a conceivable reality.
Atheroma and Cardiovascular Disease
Cardiovascular disease is a leading cause of mortality and is frequently caused by atherosclerosis.1 Atheroma formation in the aorta reflects a systemic pathology which predisposes individuals to thromboembolic or occlusive vascular events.2,3 Predicting acute vascular episodes in individuals with atherosclerosis poses a diagnostic challenge. Historically, atherosclerosis was detected by the degree of luminal stenosis on invasive angiography and extraluminal plaques remained undetected.4 Many acute thromboembolic episodes occur in patients with mild plaques following plaque rupture and the functional condition of the plaque may be an important predictor of subsequent clinical events.3,4 Optimal atherosclerosis imaging would seek to determine both anatomical severity and detect the risk of plaque rupture.
Plaque targets for molecular functional imaging in atherosclerosis
Targets of imaging
99mTc ox LDL
99mTc ox LDL-R
Annexin V micelles
Aortic imaging is a readily accessible method of identifying the presence of atherosclerosis which is increasing recognised as a systemic condition.1 Aortic atheroma is therefore more than the presence of a focal stenosis but reflects generalised vascular pathology of endothelial dysfunction and systemic inflammation. As such the identification of atheroma in one vascular bed predicts atherosclerosis elsewhere.2 Aortic atherosclerosis has been identified as surrogate marker for coronary artery disease,9-11 and the extent of aortic plaque predicts coronary events.10 Similarly, aortic atheroma is associated with cerebrovascular events and peripheral vascular disease.12-17 Aortic atherosclerosis may also produce localised complications from either thrombo-embolism or the development of adverse positive remodelling with aneurysm formation.
Aortic Imaging and the Diagnosis of Atherosclerosis
Detailed ultrasound examination of the abdominal aorta can be performed in young adults and it is possible to measure the ratio of intimal to media thickness (IMT).18-20 Aortic IMT increases with age at a greater rate that carotid IMT.18 Carotid IMT (CIMT) is an accepted maker of atherosclerosis and a predictor of cardiovascular events. National guidelines indicate that CIMT measurement is appropriate in patients with intermediate pretest risk of cardiovascular disease.21 The variation in measurements with aortic IMT was greater that that of CIMT (coefficient of variation 18% vs 3%, respectively).18 This is likely to reflect a limitation of ultrasound to achieve high resolution at increasing depths of tissue penetration and may prevent aortic IMT becoming adopted as an accepted method of cardiovascular risk assessment.
Transesophageal echocardiogram (TEE) is less susceptible to limitations in resolution as a result of the required scanning depths. It is semi-invasive and may not be tolerated in all patients. It does detect atheroma within the ascending and thoracic aorta and some features of plaque characteristics have been described for echocardiography.12 Yet in a meta-analysis the sensitive of trans-oesophageal echocardiography to detect atheroma was low and therefore in the absence of plaque detection in patients with cerebral thrombo-embolic events a further investigation is recommended to exclude aortic aetiology.22
Computed tomography (CT) detection of atheromatous aortic plaque has been used to predict vascular events23 and has been considered as a stand-alone imaging modality to determine the source of embolus in thrombo-embolic stroke.24 In comparison with multi-modality techniques contrast enhanced CT was able to find the source of cerebral events in 83% of patients.24 CT does use x-ray radiation and although in an aged population may be useful as diagnostic tool, repeated evaluations would expose individuals to a high radiation load and should be avoided where alternatives exist.
Magnetic resonance imaging (MRI) is a useful alternative to CT for anatomically defining the presence of aortic plaque and benefits from not using ionising radiation. Although the utility of MRI angiography (MRA) is improving, spatial resolution is less than with CT angiography which limits the usefulness of MRA to large calibre vessel assessment. Nevertheless, MRI contrast enhanced techniques have proven to be of value in the detection of aortic plaque and offer an opportunity to repeatedly image patients without subjecting them to the risks of radiation exposure.25
Invasive angiography is rarely used now alone in the investigation of aortic disease. This reflects the limitations of the technique to visualise complex lesions and speaks to the advances in 3D reconstruction available through CT and MRI image manipulation. Invasive techniques in addition carry inherent increased clinical risks that can be avoided with non-invasive methods, although contrast complications can be common to both approaches.
Prognosis and Identification of Unstable Plaques Using PET/CT FDG Imaging
Imaging correlates of plaque features
Non-invasive characterisation of vulnerable plaques
Lipid or haemorrhage
Lipid rich plaque
One of the first PET molecular imaging agents to be associated with atherosclerotic uptake clinically was fluorine-18 2 deoxy-d-glucose (FDG). Uptake was noted in the aorta in oncology patients undergoing total body PET/ CT imaging with FDG.28-30 FDG is a metabolic tracer that is taken up by cells as an analogue of glucose and becomes practically fixed inside the cell following phosphorylation by hexokinase.31 The rates of glucose and FDG phosphorylation in the cell are proportional to each other and reflect the general rate of glucose metabolism in the cell. FDG uptake is usually quantified as the mean or maximum standardised uptake value (SUV) which relates activity in the region of interest to the dose and the patients weight.31 As little information regarding the anatomical location of the FDG signal is provided by PET imaging it is usually combined with CT or MRI.
In preclinical32,33 and more recently in clinical studies34-37 following carotid endarterectomy, FDG uptake by plaque was seen to be correlated with plaque macrophage density. Tawakol et al34 imaged 17 patients with FDG PET/CT prior to carotid endarterectomy and found mean target to background ratio (TBR) of FDG uptake to closely correlate (r = 0.85) with mean immunohistochemistry expression of a macrophage marker (CD68). In this study, FDG uptake did not correlate with smooth muscle cell density or plaque size.
Other studies29,30,38,39 have demonstrated that FDG uptake by vessels is rarely seen in areas of vessel calcification. Such differences have prompted some investigators to comment that FDG uptake in comparison to calcification represents a different stage of the atherosclerotic disease process. It is proposed that an initial inflammatory process is later succeeded by calcification and ossification of the plaque.30 Concern has been raised however that FDG uptake may occur in calcified plaques as a result of non-specific trapping in the porous mineralised material following these observations ex vivo.40 Yet the initial impressions in vivo are that this is rarely the case given the somewhat surprising observation that overlap of FDG uptake and arterial calcification occurs infrequently (Overlap was seen in <2% of cases in one series of 78 patients Dunphy et al30).
If FDG uptake by plaques is to be considered as a functional metabolic tracer for imaging atherosclerotic disease then vascular outcome data will be required. To date few studies have examined the long-term outcomes of patients with increased vascular FDG uptake. 2 studies in oncology patients do suggest that FDG uptake may be indicative of increased risk of acute plaque events.41,42 Paulmier et al41 cross-matched 45 patients with FDG uptake with 56 patients having no vascular FDG uptake, controlling for major clinical risk factors. Plaque FDG uptake was seen to predict recent cardiovascular events, whereas vessel calcification predicted the occurrence of previous events. Rominger et al42 followed 932 oncology patients without prior cardiovascular disease who had undergone an FDG PET/CT study. Patients whom experienced subsequent cardiovascular events had higher baseline vascular FDG uptake which appeared to be an independent predictor of subsequent cardiovascular episodes.
These promising results suggest FDG uptake by culprit lesions might be identifiable by PET and could help direct clinical intervention to appropriate vessels. No standard methods have been proposed, however, to robustly identify culprit lesions. In addition, different protocols between studies make inter-study comparisons difficult. More uniform methodology has been called for to better develop atherosclerosis imaging with FDG PET.43
Other approaches with PET radiopharmaceuticals have been used and include choline-based 11-Carbon (11C-choline) and 18F (18fluorocholine) tracers.44,45 Choline is a source of cell membrane lipids and vessel wall choline based tracers co-localize with plaque macrophages and inflammation in animal models of atherosclerosis.46,47 PET/CT choline tracers are used in the investigation of prostate cancer patients on account of the high phosphatidylcholine turnover in these cells.48 Clinical studies have been performed to determine whether vessel wall choline uptake might describe the extent and associated risk from aortic atherosclerosis. In 93 prostate cancer patients undergoing 11C-choline PET/CT imaging the uptake of radiotracer in the aorta was compared to the distribution of atheromatous calcification identified by CT.44 Similar to FDG imaging little co-localization was observed between choline uptake and vessel calcification (only 1% of calcified regions demonstrated trace uptake). A different group were also able to demonstrate vessel 18F-choline uptake in prostate cancer patients and again there was little colocalization with vessel calcification.45 In this group of 60 patients although vessel calcification was correlated with clinical cardiovascular risks, vessel 18F-choline uptake was not. Further corroborative evidence is required, therefore, to determine the clinical relevance of choline uptake in vessels.
Emerging Functional Molecular Markers of Vulnerable Plaques
Animal studies using micro-PET and SPECT with CT have evaluated a number of molecular imaging biomarkers of plaque vulnerability (Figure 1, 2, 3). Plaque targets for these novel tracers have included oxidized LDL receptors,51 markers of plaque inflammation (MTP-1)52 or platelet activity (ADP analogues),53 apoptosis (Annexin-V)54 and plaque angiogenesis (integrins)55 attached to 99m-Technetium (99m Tc) or FDG, for SPECT or PET imaging, respectively. Further studies consistent with the clinical data presented have confirmed the uptake of FDG uptake in atherosclerotic lesions and demonstrated the correlation between FDG uptake and macrophage density within lesions.32,33
Plaque angiogenesis has also been used as a biomarker in contrast enhanced ultrasound (CEU) detection of atherosclerosis in animal studies. Verification of the promise of this technology56 was produced recently when CEU was performed using contrast labelled with vascular endothelial factor receptor antibodies.57 Anti-VEGFR-contrast increased the sensitivity of this technique to detect vessels within atherosclerotic lesions which can be a feature of unstable plaques.7,8,58,59
Progress is also being made in identifying suitable MR and CT functional markers of plaque phenotype. Super oxide paramagnetic iron nanoparticles are taken up by macrophages in atherosclerotic lesions. Preclinically this has been used successfully in animal studies to document plaque macrophage content and plaque inflammation.60 Other nanoparticles to integrins (alpha nu beta 3 (ανβ3)), matrix metalloproteinases (MMPs) and micelles labelled with macrophage receptors or products (neutrophil gelatinase associated lipocalin (NGAL), peripheral cannabinoid receptor- (CBR-2)) have also successfully identified plaques with vulnerable traits in vivo studies.61-64 Ultra-small paramagnetic iron oxide (USPIO) nanoparticles have also been assessed in a pilot study in patients with aortic aneurysm. Uptake of USPIO was associated with macrophage infiltration confirmed by histology following aneurysm repair.65
Fewer novel CT contrast agents are being developed to assist in the differentiation of quiescent from liable lesions. In murine studies, gold nanoparticles targeted to HDL have been used in preclinical spectral CT systems to identify macrophage rich plaques.66 In this system, incident x-rays are divided into different energy bins for multicolour imaging. This system promises to be useful tool in atherosclerotic imaging since it combines both calcium- and iodine-based techniques with the ability to acquire functional plaque data using atherosclerosis-directed nanoparticles.66
Clinical and preclinical data are awaited that confirm the optimism associated with functional molecular imaging. Although retrospective data is available that suggests a link between functional imaging data and clinical outcomes, more prospective prognostic data are required.
Monitoring Response to Therapy Using PET/CT Imaging
Accompanying developments in diagnostic atherosclerotic imaging, investigators have also used modern imaging techniques to determine the effectiveness of anti-plaque therapy. Serial imaging is more attractive with non-invasive techniques particularly in asymptomatic individuals where invasive testing is difficult to justify.
The efficacy of HMG-Co-A reductase inhibitors (statins) to reduce plaque inflammation was assessed in 43 patients with high baseline FDG uptake in aorta and/or carotid arteries. Individuals that received simvastatin (5-20 mg/day) demonstrated a 10% reduction in mean standardized uptake values (SUV) while there was no change observed in the control group who received dietary advice.67 Similarly, FDG vessel uptake was assessed in 60 healthy individuals in whom 83% had FDG uptake in at least one vascular bed. Following lifestyle changes that resulted in reduce LDL cholesterol and raised HDL cholesterol a decrease in the number of sites of FDG uptake was seen.68
MRI has also been used to monitor the effect of therapy on plaque dynamics. Patients with familial hypercholesterolemia were followed using MRI to determine whether intensive cholesterol lowering therapy would reduce aortic plaque burden.69 Similarly, MRI has been used to evaluate whether plaque regression responded to low (simvastatin 20 mg) or high dose (simvastatin 80 mg) therapy70 or treatment with fibrates.71
Whether MRI or FDG PET is used in future to monitor responses to therapy may depend upon the population studied and the evaluation required. The absence of radiation exposure with MRI is an advantage particularly in young individuals in whom serial scans are planned. FDG-PET/CT, however, may provide useful data regarding the response to anti-inflammatory therapy that may not be available using MRI and although radiation exposure is a concern the doses associated with FDG-PET imaging can be reasonably low at 6.4 mSv.72
In animal studies, functional molecular imaging has been employed to deliver targeted anti-atherogenic therapy. Using alpha (nu) beta3 integrin-targeted nanoparticles, atheroma formation was reduced in plaque prone rabbits by incorporation of fumagillin (an angiogenesis inhibitor) into the contrast moiety.73 Rabbits treated with fumagillin had smaller plaques visible at MRI and fewer microvessels within the plaques assessed histologically. Integrin-targeted SPECT and PET tracers are also available55,74 and offer radionuclide opportunities therefore to deliver and monitor the response to targeted anti-atherogenic therapies in pre-clinical studies.
Aortic atheroma is both a localized marker of potential pathology and an indicator of systemic atherosclerosis.1,3 Future clinical directions for diagnostic imaging of the aorta are likely to reflect both these aspects. As a consequence of the local manifestations of atherosclerosis in the aorta, imaging of the aorta could be a useful non-invasive method of risk stratification. In time aortic assessment may become an acceptable method to further assess patients, mirroring the acceptance of carotid IMT in this setting.21 Detection of aortic inflammation on the other hand by nuclear diagnostic imaging may provide pertinent information concerning either downstream events or events in upstream vascular beds. In this setting, radionuclide functional imaging of the aorta could provide prognostic data to help risk stratify patients acutely and identify “vulnerable patients.” Potentially, these techniques might be useful as diagnostic tools therefore in the management of acute chest pain syndromes or transient ischaemic cerebral events.
Despite a reduction in cardiovascular disease in recent years and an improvement in acute and preventative therapies, there is still a clinical need for improved diagnostic imaging in atherosclerosis. Sudden vascular events remain largely unheralded1 and clinical treatment of patients with atherosclerotic plaque is based upon global risk reduction not plaque dynamics. Management of symptomatic individuals is often decided upon from anatomical data from luminography which does not accurately predict the recurrence of events.1,75,76 To solve some of these issues clinicians and imaging physicians have sought to assess both plaque anatomy and function. Ultrasound and CT and to a lesser extent MRI have not yet provided reliable data concerning plaque function and the aspirations of imagers have seemed a distant hope.
Into this setting, however, nuclear imaging techniques have emerged with data identifying inflammatory components within plaques. Combined FDG PET/CT imaging has begun to retire the notion that such aspirations for imaging were ethereal. Instead functional molecular imaging with radionuclides and other contrast agents is poised to translate much of the promise observed in preclinical studies to help better identify unstable lesions. In the near future, the arrival of this technology will alter the way we screen, manage and treat patients with atherosclerotic disease.
- 15.Atherosclerotic disease of the aortic arch as a risk factor for recurrent ischemic stroke. The French Study of Aortic Plaques in Stroke Group. N Engl J Med 1996;334:1216-21.Google Scholar
- 38.Rudd JH, Myers KS, Bansilal S, Machac J, Woodward M, Fuster V, et al. Relationships among regional arterial inflammation, calcification, risk factors, and biomarkers: A prospective fluorodeoxyglucose positron-emission tomography/computed tomography imaging study. Circ Cardiovasc Imaging 2009;2:107-15.PubMedCrossRefGoogle Scholar
- 48.Krause BJ, Souvatzoglou M, Treiber U. Imaging of prostate cancer with PET/CT and radioactively labeled choline derivates. Urol Oncol 2011.Google Scholar
- 62.Hyafil F, Vucic E, Cornily JC, Sharma R, Amirbekian V, Blackwell F, et al. Monitoring of arterial wall remodelling in atherosclerotic rabbits with a magnetic resonance imaging contrast agent binding to matrix metalloproteinases. Eur Heart J 2010.Google Scholar
- 63.Olzinski AR, Turner GH, Bernard RE, Karr H, Cornejo CA, Aravindhan K, et al. Pharmacological inhibition of C-C chemokine receptor 2 decreases macrophage infiltration in the aortic root of the human C-C chemokine receptor 2/apolipoprotein E−/− mouse: Magnetic resonance imaging assessment. Arterioscler Thromb Vasc Biol 2010;30:253-9.PubMedCrossRefGoogle Scholar
- 64.te Boekhorst BC, Bovens SM, Rodrigues-Feo J, Sanders HM, van de Kolk CW, de Kroon AI, et al. Characterization and in vitro and in vivo testing of CB2-receptor- and NGAL-targeted paramagnetic micelles for molecular MRI of vulnerable atherosclerotic plaque. Mol Imaging Biol 2010;12:635-51.CrossRefGoogle Scholar
- 65.Richards JM, Semple SI, Macgillivray TJ, Gray C, Langrish JP, Williams M, et al. Abdominal Aortic Aneurysm Growth Predicted by Uptake of Ultrasmall Superparamagnetic Particles of Iron Oxide: A Pilot Study. Circ Cardiovasc Imaging 2011.Google Scholar
- 69.Schmitz SA, O’Regan DP, Fitzpatrick J, Neuwirth C, Potter E, Tosi I, et al. Quantitative 3T MR imaging of the descending thoracic aorta: Patients with familial hypercholesterolemia have an increased aortic plaque burden despite long-term lipid-lowering therapy. J Vasc Interv Radiol 2008;19:1403-8.PubMedCrossRefGoogle Scholar
- 70.Corti R, Fuster V, Fayad ZA, Worthley SG, Helft G, Chaplin WF, et al. Effects of aggressive versus conventional lipid-lowering therapy by simvastatin on human atherosclerotic lesions: A prospective, randomized, double-blind trial with high-resolution magnetic resonance imaging. J Am Coll Cardiol 2005;46:106-12.PubMedCrossRefGoogle Scholar
- 77.Hartung D, Petrov A, Haider N, Fujimoto S, Blankenberg F, Fujimoto A, et al. Radiolabeled Monocyte Chemotactic Protein 1 for the detection of inflammation in experimental atherosclerosis. J Nucl Med 2007;48(11):1816–21.Google Scholar