Introduction
Ischemic heart disease is the leading cause of death worldwide (1). Understanding cellular alterations involved in atherosclerotic plaque instability, as well as reparative mechanisms following myocardial infarction (MI) is important for both preventive and therapeutic intervention.
Positron emission tomography (PET) is an in vivo medical imaging technique that enables quantification of radiotracer uptake in the entire body at a cellular level.
Integrin αvβ3 is a transmembrane receptor mediating cell adhesion that influences cell growth, proliferation, survival, and migration. The peptide motif arginine–glycine–aspartate, abbreviated by “RGD” in the one-letter code, has been identified in 1984 as a minimal amino acid sequence that some integrins recognize in their natural ligands (2). In 1991 a group from Germany reported a strong and selective binding of cyclic pentapeptides containing the RGD sequence to αvβ3 integrin (3). The past 20 years have witnessed a remarkable expansion on cyclic peptides containing the RGD sequence since it was found that αvβ3 integrin plays a major role in angiogenesis, further enhanced by the first successful clinical applications of the PET radiopharmaceutical 18F-Galacto-RGD (4–8).
The integrin-mediated cell adhesion/migration imaging with RGD PET was found to be involved in a wide range of pathophysiological processes, not only related to angiogenesis. Increased αvβ3 expression has been observed in some cancer cells as well as in cells involved in extracellular matrix remodeling such as fibroblasts and activated macrophages (8, 9). However, the clinical relevance of such biomarker is not well understood (8). Although most studies using RGD-PET agents focused predominantly on tumorigenesis, a greater focus on non-oncological applications such as cardiovascular diseases is of interest (9). In fact, in the development of atherosclerosis, expression of integrin αvβ3 has been found in endothelial cells as well as in CD68-positive macrophages, which are known as key factors in plaque instability (10–12). Moreover, αvβ3 integrin expression seems to appear central to the coordination of myocardial repair following MI. αvβ3 integrin is indeed upregulated in states of angiogenesis within the infarcted myocardium and could be expressed by activated myofibroblasts and macrophages during margination and chemotaxis (13, 14).
Some recent studies using RGD-PET agents in cardiovascular diseases have been published. The aim of this systematic review is to perform an updated evidence-based summary about the role of RGD-based PET agents in patients with cardiovascular diseases to better address further research in this setting.
Discussion
Since the first successful clinical applications of an RGD PET radiopharmaceutical two decades ago, all the final eligible articles included in this review (n = 7) were published in the last 9 years, reflecting a recent growing interest in RGD PET/CT imaging focused on cardiovascular diseases (16–22). These studies varied in size and methodology. Two main topics can be distinguished: the infarcted myocardium and atherosclerosis. Stroke is a third topic that can be distinguished, but slightly less studied. All studies had a relatively clearly stated, clinically relevant, and patient-related purpose. Longitudinal studies that tell more about the natural history of αvβ3 expression in cardiovascular diseases using PET imaging were in very short supply. Interventional studies or randomized controlled trials have not been conducted so far. Due to this diversity, our systematic review could not provide an answer to a particular scientific question; we were left to summarize the main information about RGD PET in cardiovascular diseases that the literature had provided until now and point to some major issues that remain unanswered.
Infarcted Myocardium and α<sub>v</sub>β<sub>3</sub> Integrin Expression
Ventricular remodeling defines the changes that occur in structure, geometry, and function of the myocardium after MI. This biological process involves inflammation, angiogenesis, repair, and healing with specific biochemical and structural alterations in the infarcted myocardium, peri infarcted and remote regions. Several cell populations are involved, such as neutrophils, monocytes/macrophages, fibroblasts, T cells, stem cells, etc. (23, 24). Understanding reparative mechanisms following MI is becoming increasingly important. In some circumstances, maladaptive persistent processes may be detrimental. This may encourage remodeling and scarring to extend into the myocardium long after the initial causative injury, leading to progressive ventricular dilatation, ventricular dysfunction, and heart failure (25, 26).
The expression of αvβ3 integrin following MI, using PET imaging technique, could be of great interest. In contrast to a low level of expression by quiescent endothelial cells, αvβ3 integrin was found to be upregulated in state of angiogenesis within the myocardium after infarction in preclinical studies in rats (13, 14). αvβ3 integrin expression was also documented by both activated cardiac myofibroblasts and macrophages during margination and chemotaxis. Thus, αvβ3 integrin expression could play a central role in the coordination of the repair processes following MI.
Prevalence and Natural History of RGD PET Uptake in the Infarcted Myocardium, and Future Potential Directions
Forty-six out of the 56 (90%) MI patients in PET studies focused on the infarcted myocardium presented RGD uptake in the infarcted area, demonstrating that αvβ3 integrin expression in the infarcted myocardium is well evident in RGD PET/CT scans.
In human studies of acute MI, RGD-based PET radiotracers accumulate at the site of infarction as early as 3 days and seem to be peaking at 1–3 weeks post-MI before being decreasing (16, 18). However, only one study assessed serial changes of myocardial RGD uptake after ischemic intervention (18). Moreover, the timepoint of the second 18F-fluciclatide PET/CT after MI in this single study was particularly variable (76 ± 19 days). Changes of myocardial RGD PET uptake in humans after coronary occlusion and reperfusion is still to be precisely determined. Serial changes of myocardial RGD PET uptake are studied in more detail in an ongoing prospective study (ClinicalTrial.gov NCT03809689).
Vessel damage with ischemic insult is expected to occur during the first days after MI (27, 28), which could lead to unspecific tracer diffusion in damaged leaky vessels. Here, the peak of tracer accumulation at 1–3 weeks post-MI is in favor of minimal effects of this unspecific tracer diffusion in damaged leaky vessels. Myocardial uptake post MI using RGD PET/CT may thus be predominantly due to integrin expression rather than unspecific leakage. However, no imaging was performed at day 1 in the different studies.
An interesting finding in both studies from Sun et al. (16) and Jenkins et al. (18) was the absence of increase in RGD uptake in chronically damaged myocardium, indicating minimal residual αvβ3 integrin expression in old ischemic injuries. Therefore, αvβ3 integrin expression seems rather to act as a marker of cardiac repair at sites of recent MI (16, 18).
Despite those promising results further prospective studies are nevertheless warranted to confirm a clinical value of RGD PET uptake as a prognostic marker for left ventricular remodeling. Moreover, some findings support increased αvβ3 expression also in non-viable tissue, which could be explain by a linking to the myofibroblasts (16, 29). Studies assessing the prognostic value of RGD PET uptake in the infarcted myocardium are still needed.
Atherosclerosis and α<sub>v</sub>β<sub>3</sub> Integrin Expression
The ability to accurately and non-invasively monitor inflammatory processes in atherosclerotic plaques would represent a major advance, and metabolic imaging using PET has the potential to fulfill this important unmet clinical need. Integrin αvβ3 has been widely studied in atherosclerosis. In the development of atherosclerosis, expression of integrin αvβ3 has been found in endothelial cells as well as in CD68-positive macrophages, and both inflammation and angiogenesis processes are associated with plaque growth, plaque instability and clinical events (10–12). The necrotic core in the culprit plaque results from increasing inflammation (30). Pro-atherogenic stimuli lead to an infiltration of activated monocytes within the intima, which differentiate into pro-inflammatory macrophages (31). The apoptosis of the resident pro-inflammatory macrophages leads to the development of a lipid-rich or necrotic core (30). Matrix metalloproteinases secreted by macrophages weaken the fibrous cap, predisposing it to rupture. Angiogenesis is believed to occur in response to hypoxic conditions within the necrotic core. Neovessels, arising from the adventitial vasa vasorum, grow into the base of progressive atherosclerotic lesions and provide an alternative entry pathway for monocytes and immune cells. The plaque neovessels are fragile and leaky, giving rise to local extravasation of plasma proteins and erythrocytes (32). Plaque hemorrhage itself results in a pro-inflammatory response, plaque destabilization and clinical events (11, 33).
Attractive Promising Results of RGD-Based PET Agents in Atherosclerosis
Overall, although the numbers are limited, the existing studies support the potential of RGD-based PET imaging of αvβ3 integrin expression in atherosclerosis. 18F-fluciclatide PET/CT has proven itself as a potential valuable target for the imaging of unstable atherosclerosis (19). 68Ga-NODAGA-RGD uptake in large vessels showed clinical correlation with atherosclerotic cardiovascular diseases (21). 18F-Galacto-RGD PET/CT showed specific tracer accumulation in human atherosclerotic carotid plaques, with increased PET signals observed in stenotic compared with non-stenotic areas based on MR or CT angiography data, and with histological correlation with αvβ3 expression (17).
Intensity of RGD PET Uptake in Plaque Imaging in Humans
Background corrections were systematically used among different studies. Again, the median intensity of arterial tracer uptake as determined by TBR measurements were 1.31 (IQR 1.20–1.39) for the study from Jenkins et al. (19) and 1.84 (IQR 1.62–2.04) for the study from Dietz et al. (21). Interestingly, Dietz et al. used a SiPM PET system (Biograph Vision 600, Siemens Medical Solutions, Knoxville, USA) in 16 out of 44 (36%) of the patients included for the analysis, which could partly explain the highest TBR measurements. SiPM PET is a new PET technology with improved spatial and timing resolution and a relatively high sensitivity and count-rate capability as compared to PET scanners using conventional photomultiplier tubes (34).
The TBR measurements were slightly lower than those of other established tracers used for plaque imaging. In a study using 18F-sodium fluoride for visualization of microcalcification in plaque in the major arteries, the mean TBR was 2.3 ± 0.7 (35). In another study using 68Ga-Pentixafor for visualization of inflammatory cells, the mean TBR was 2.0 ± 0.5 (36).
The difference in intensity of TBR between non-calcified/mixed/calcified plaques on CT has not been studied so far. However, Dietz et al. found that only around 26% of 68Ga-NODAGA-RGD highest uptake foci in their cohort were colocalized to calcification on CT (21). This result suggests that RGD accumulation could particularly occur mainly in the non-calcified vessel wall, which may indicate some association with pathophysiologic processes found in early atherosclerotic disease.
Future Potential Directions of α<sub>v</sub>β<sub>3</sub> Integrin Expression in Plaque Imaging
To date, RGD PET imaging in humans has emerged as a promising tool for the assessment of active atherosclerosis processes, but with a still very limited number of studies (n = 3) (17, 19, 21). All studies were focused on subjects with established disease, and there is no work which has investigated whether RGD uptake could predict future events or whether it can provide additional prognostic information over current methods of risk stratification. Further works are also needed to establish whether RGD PET could monitor disease progression, guide therapeutic interventions, or assess novel anti-angiogenic therapies in cardiovascular diseases.
It is also not well established whether having a recent MI would lead to a significant increase in aortic plaque microvessels, or indeed whether these changes are likely to occur within 1–3 weeks post MI. Because integrins αvβ3 are expressed in neovessels in plaques but also in CD68-positive macrophages, it is not possible to ascertain with PET only whether RGD is binding preferentially to one or the other of these processes. Aortic RGD uptake post MI in large vessels could be reporting predominantly on systemic inflammation rather than microvessels proliferation (37).
No study assessed the coronary arteries. The potential value of integrin tracers would be of great interest for future trials, as most PET tracers are not suitable due to too low signal for coronary artery disease assessment. Moreover, the capacity of the paradigm based on ischemia to prevent MI is increasingly being questioned, and the quantification of metabolic activity of atherosclerotic disease is of growing interest (38). PET imaging with integrin tracers could have the potential to add value to this purpose.
No PET study in humans assessed αvβ3 integrin expression in aortic aneurysm, which could be of clinical interest. Degraded aortic tissue is known to exhibit abnormal angiogenesis and is present in abundance at sites of aortic rupture (39). αvβ3 integrin expression stimulate angiogenesis, typically in hypoxic environments (40). The predictive value of αvβ3 integrin expression imaging using PET in aortopathy to predict disease progression is unknown.
RGD PET Uptake in Stroke
PET studies focused on ischemic cerebral vascular disease are scarce (n = 2) (16, 22). As in the normal myocardium, the PET RGD-peptide was absent in the normal brain, allowing a clear background to evaluate the post-stroke angiogenesis. However, the RGD PET uptake levels in stroke seem to be lower than those in MI (16). These could indicate difficulties of forming new capillaries from the pre-existing vessels after stroke (16). The existing data support that αvβ3 integrin expression in the cerebral infarct is prominent in the acute phase (<30 days after the event) and decrease with time, but evidence is still lacking. Serial changes of cerebral RGD uptake were assessed for only two patients (16). However, both existing studies have consistent findings in chronically damaged brain, with absent or very weak uptake reported.
Some evidence supports a strong correlation between RGD uptake and stroke size but only in a limited number of patients scanned 13–26 days after the event (16). Like in MI, a non-perfect correlation between RGD uptake and standard measures of infarct severity like the infarct size could suggest that RGD uptake is not a surrogate of infarction but could relate more to the tissue-healing response to injury.
An interesting finding in the study from Sun et al. is a punctate multifocal RGD uptake along the surrounding blood vessels in stroke patients, contrasting with the patchy-form uptake found in MI patients (16). The authors explained that the myocardial remodeling could contribute to this different uptake pattern, but more data are needed to confirm these findings.
Radiolabeling of RGD-Based PET Agents
Most of the RGD-based radiotracers included in this systematic review were labeled with 18F (4 out of the 7 studies: 57%). The remaining 2 out of the 7 studies (29%) used RGD-based radiotracers labeled with 68Ga obtained from a 68Ge/68Ga generator. Imaging relatively thick structure like the left ventricle is not an issue using PET imaging but the use of 18F labeled tracers for atherosclerotic PET imaging could have several physical and technical advantages as compared to 68Ga labeled tracers. In fact, 68Ga has higher positron energy and positron range which could lead to noisier images and worse spatial resolution, which could be amplified by a lower injected activity of 68Ga compared with 18F (at least using traditional photomultiplier tube PET/CT). However, the synthesis of 18F-Galacto-RGD tracer used by Beer et al. could be complex and time consuming, making an automated production process, which is mandatory for routine clinical use, extremely difficult (17, 41, 42). Moreover, 68Ga labeled tracers have already been extensively used to image atherosclerotic plaques in large vessels, and in vessels as small as the coronary arteries (36, 43, 44). The superiority of a labeling with 18F or 68Ga is not clear for RGD-based radiotracers for cardiovascular PET imaging.
Limitations
We should report several limitations in our systematic review, which could limit the scope of our results. Our eligibility criteria lead to a very small number of original articles eligible for inclusion in this systematic review, with limited histological assessment. This systematic review focused on PET, and studies focusing on SPECT were not included. Unlike SPECT imaging, PET imaging does not require an extrinsic collimator resulting in higher count sensitivity and spatial resolution. Moreover, accurate and well validated attenuation correction is available with PET. This provides routinely available quantitative assessments (45), which could be crucial. In fact, important potential clinical applications of molecular integrin imaging in cardiovascular diseases could provide quantitative endpoints for use in clinical trials.
There was significant heterogeneity across the studies in terms of technical aspects. Different radiotracers were used. The radiopharmaceutical injected activity and the mean time interval between radiotracer injection and image acquisition varied among the included studies.
For atherosclerosis imaging, it cannot be ruled out that tracer signals observed with RGD PET are at least partially attributable to noise, especially using 68Ga labeling (46).
Finally, despite a high number of prospective studies with relatively clearly stated and patient-related purposes, according to the NIH quality assessment tool the quality of the seven selected studies is judged as moderate, mainly as they were single center and with a limited number of patients. We have tried to minimize publication bias excluding case reports or small case series from this systematic review.