Introduction
In patients with coronary artery disease (CAD), inflamed and prone-to-rupture coronary plaques are associated with higher risk of major adverse cardiac events (MACE) [
1]. Accordingly, numerous invasive and non-invasive approaches for their early identification and characterization have been tested. Among various possible imaging targets, macrophage infiltration, especially sustained by pro-inflammatory monocyte-derived macrophages (M1-phenotype), have emerged as a potential marker of plaque vulnerability [
1,
2]. In fact, lipid-derived metabolites such as low-density lipoprotein (LDL), that can be found in coronary plaques, are known to stimulate the migration of macrophages into the arterial intima, wherein they mature and become phagocytic [
2,
3]. These “activated” macrophages then upregulate inflammatory metabolic pathways causing the progression of CAD [
1,
4].
Positron emission tomography (PET) is an excellent tool for the assessment and the characterization of various metabolic processes. Metabolic radiotracers like [
18F]FDG proved reliable in the evaluation of inflamed plaque [
5], but lacks specificity for the identification of activated macrophages. In this regard, somatostatin receptor 2 (SST
2) imaging may represent an important advance, as SST
2, a G-protein-coupled transmembrane protein, is overexpressed by M1-macrophages [
6,
7]. Consistent with this concept, it was reported that SST
2 imaging yields improved accuracy in discriminating high-risk versus low-risk coronary lesions than [
18F]FDG [
8,
9].
While SST
2 imaging with PET has proven promising in the evaluation of inflamed vascular plaque in patients investigated for oncological reasons [
10,
11], its wider implementation has been precluded by its lower diagnostic accuracy compared to [
18F]FDG as a result of its lower sensitivity. The relatively low total number of activated macrophages in plaques causes low signal-to-noise ratio (SNR) in SST
2 imaging, whereas [
18F]FDG is imported by glucose transporters (GLUT), which are upregulated on a wide variety of inflammatory cells and not limited to macrophages [
12,
13]. Other potential challenges include motion artifacts (cardiac and respiratory), shorter half-life (68 min vs. 110 min) and lower positron yield of the
68 Ga compared to for example
18F [
14]. Hence, areas affected by an infiltration of M1-macrophages are difficult to image on conventional PET scanners due to low signal collection efficiency and limited resolution [
15,
16]. This gap has now been closed with the introduction of new silicon photomultiplier (SiPM)-based, long-axial field-of-view (LAFOV) PET/CT scanners. The recent clinical implementation of LAFOV PET/CT with 15 fold improvement in sensitivity and a spatial resolution of approximately 3 mm allows for identification and quantification of small areas with low radiotracer uptake [
17‐
21]. The higher sensitivity of LAFOV systems results in higher temporal resolution [
22]. This could be of utility in gated acquisitions and facilitate the imaging of structures vulnerable to motion artifacts, such as the coronary arteries.
This study aims to evaluate the detectability of calcified coronary artery plaques overexpressing SST2 on LAVOF PET scanners. To investigate SST2 overexpression as marker of plaque vulnerability, PET findings were correlated to cardiovascular risk factors and clinical outcomes.
Discussion
In this study, we report the first data for [
68 Ga]Ga-DOTA-TOC imaging to detect inflamed calcified coronary artery plaques using a LAFOV PET/CT system. Previous work indicated that PET-based SST
2 imaging has a clear potential in the detection of inflamed vascular plaques, with a more specific affinity compared to [
18F]FDG [
8,
9], but to-date, the detection of SST
2 positive calcified plaques in routine PET/CT has been challenging. Li et al. report increased [
68 Ga]Ga-DOTA-TATE uptake in 4/16 patients with calcified plaques within large arteries, resulting in a detection rate of 25% [
31]. The detection and quantification of [
68 Ga]Ga-DOTA-TOC uptake in small vessels such as the coronary arteries has been even more challenging due to intrinsic, relative limitations in spatial and temporal resolution of contemporary standard-axial field-of-view (SAFOV) PET/CT systems. Such analog PET-systems with photomultiplier tubes (PMT) have lower contrast-to-noise ratios (CNR) and inferior time-of-flight resolution compared to new SiPM-based PET-systems [
15,
32]. The introduction of digital whole-body PET/CT systems like LAFOV PET/CT scanners has important advantages compared to previous generation scanners [
33]. SiPM-based LAFOV systems have improved sensitivity, higher signal-to-noise ratios and allow for a more precise localization of small target lesions, which is expected to translate into a higher detection rate of small lesions compared to standard PET scanners [
19,
26].
In this regard, we here report a patient-based sensitivity for [
68 Ga]Ga-DOTA-TOC in calcified coronary arteries of 49% with LAFOV PET/CT system. This value is higher compared to that reported by Li et al. [
31], wherein also larger arteries were investigated. Differently from previous reports, we observed significant higher [
68 Ga]Ga-DOTA-TOC uptake in patients with calcifications in more than one vessel. Moreover, increased [
68 Ga]Ga-DOTA-TOC uptake in coronary arteries correlated with the global- and LAD- CACS. In this context, we report higher TBR compared to the previous report by Rominger et al. (our TBR: 1.65 ± 0.53; Rominger et al.: 1.21 ± 0.30);, which may reflect the noise reduction and increased sensitivity of a LAFOV PET system [
9,
10,
26,
34]. However, also differences in the clinical status of their patients may have impacted these results, and this may also explain why TBR in our study was generally lower than reported by Mojtahedi et al. (2.04 ± 1.76). It should be noted that none of our patients had known CAD, while 9.1% of patients in the study by Mojtahedi et al. had prior history of revascularization.
Another important difference to previous reports is that we focused on the detection of increased DOTA-TOC uptake in calcified coronary plaques only. While this choice reflects the retrospective nature of the present study (only a low-dose CT and no CT-based coronary angiography was available), still inflamed plaques with macrocalcifications are important to detect. Previous report showed that the most frequent increased uptake of SST
2-tracers occurs in calcified plaques. Rominger et al. and Mojtahedi et al. reported that increased uptake was present in 75% of coronary artery calcifications as well as in plaques with high density (> 71 HU) [
9,
10]. Likewise, Malmberg et al. showed that high CACS is an independent predictor of increased SUV
max with [
64Cu]Cu-DOTA-TATE [
2]. Not less important, plaque calcification is a marker of atherosclerosis, and higher CACS is widely recognized as a robust predictor of MACE [
35,
36]. It should be noted that the presence of calcifications in a vascular plaque is a prerequisite for its vulnerability, and there is still a contention regarding the pattern of calcification predictive of higher risk of rupture [
37]. Although spotty calcifications have been reported as a potential risk factor for the development of higher degree of inflammation [
38], inconsistent findings were seen with regard to largely calcified plaques. Some papers demonstrated that plaque calcification was higher in asymptomatic patients than in symptomatic patients [
39], other works showed that a larger calcification volume was associated with a higher prevalence of intra-plaque hemorrhage [
40]. Of note, studies specifically investigating the role of a different calcification pattern in coronary plaques with increased uptake of SST
2-tracers are missing.
Our work expands on the association between calcified plaques with increased uptake of SST
2-ligands and cardiovascular risk. Stroke occurred more often in patients with [
68 Ga]Ga-DOTA-TOC uptake in the calcified coronary plaques than in patients without detectable uptake (
p < 0.01). More importantly, patients with [
68 Ga]Ga-DOTA-TOC avid calcified plaques had higher rate of all-cause death compared to patients without [
68 Ga]Ga-DOTA-TOC avid calcified plaques. Our data are in line with prior observations based on the evaluation of both calcified and non-calcified plaques, wherein increased uptake of DOTA-TATE within a coronary plaque was associated with higher rate of MACE independently from other established risk factors [
9,
10,
41]. The fact that a similar predictive value also applies to calcified, possibly inflamed plaques in a medium-term follow-up gives confidence to also consider inflamed calcified plaques as determinants of cardiovascular risk. Of note, although all-cause death was used as surrogate for MACE, none of the patients in the study died from oncological reasons, and although a precise cause could not be identified in 9/12 patients, a cardiac origin for the death cannot be ruled out.
In the patients undergoing PRRT, we identified a decrease in [
68 Ga]Ga-DOTA-TOC uptake within the calcified coronary plaques post-therapy (SUV
max: 1.46 ± 0.14 pre vs. 0.94 ± 0.19 post,
p: 0.01). This finding is consistent with previous reports. Schatka et al. also showed that [
68 Ga]Ga-DOTA-TATE uptake decreases after the PRRT in large vessels [
42]. However, we are now able to present first data on the effect of PPRT on vessels as small as coronary arteries. This confirms that the higher sensitivity of LAFOV PET/CT allows for detecting small changes in [
68 Ga]Ga-DOTA-TOC uptake even in small vessels. Having in mind that [
68 Ga]Ga-DOTA-TOC uptake is a marker for macrophage activity, we can assume that PRRT may reduce the degree of plaque inflammation [
42]. As such, [
68 Ga]Ga-DOTA-TOC LAFOV PET/CT might be able to identify changes of the plaque inflammation even in small vessels and might be useful to monitor anti-inflammatory therapy.
Since the presence of increased [
68 Ga]Ga-DOTA-TOC uptake correlates with a worse clinical outcome the degree of uptake may represent the degree of inflammation rather than unspecific activity. As such, we postulate that LAFOV PET/CT might afford the detection of prognostic relevant inflammatory changes in vivo [
43]. Reduced tracer uptake after PRRT may also suggest a reduction in the activity of plaque inflammation. While this may serve as a hypothetical therapeutic optional in patients with high cardiovascular risks, it also raises the notion that LAFOV-based [
68 Ga]Ga-DOTA-TOC PET/CT might serve as a tool for the monitoring of other cardiac therapies. Further studies investigating the influence of PRRT on vascular inflammation and as a tool for therapeutic monitoring are warranted.
Some limitations of our study should be acknowledged. First, as LAFOV systems were recently introduced, our patient sample is small. At the time of investigation, neither ECG-triggered acquisitions nor algorithms to correct for motion artifacts were available for this scanner. Therefore, CT-based coronary angiography (CCTA) was not performed. Thus, as previously mentioned, the retrospective nature of our study prevented us to evaluate the impact of [
68 Ga]Ga-DOTA-TOC uptake in non-calcified plaques. Additionally, we included oncologic patients referred for a [
68 Ga]Ga-DOTA-TOC PET/CT and did not select patients with CAD, who might have been treated by a cardiologist in the follow-up period. However, our cohort represents a real world setting where the aim is to characterize cardiac lesions of risk for inflammation as soon as possible. In this regard, it should be noted that patients were not on oncologic therapies other than somatostatin analogs, which exclude a potential bias due to therapeutic regimen. Furthermore, the fact those patients were also not on cardiologic therapy and did not change their therapeutic regimen after PET excludes another potential bias in the evaluation of the prognosis. Most calcified plaques were located in the LAD. The frequency of calcified plaques with [
68 Ga]Ga-DOTA-TOC uptake apart from LAD was low. Therefore, the prognostic impact of lesion location could not be properly assessed. This is a limitation of our data and should be addressed in further studies. However, the correct identification of significant stenosis (potentially caused by inflamed plaques) in the LAD represents a paramount of importance for therapy decision. In fact, a > 50% stenosis of the proximal LAD with evidence of ischemia is currently considered a robust indication for a successful revascularization [
44,
45]. The fact that no correlation between cardiovascular risk factors and the degree of uptake was found differs from what reported in the previous studies [
9,
10]. The fact that a different camera system was used may partly explain this discrepancy, as well as the fact that our population did not consist of patients with CAD. But an explanation of this aberrance requires further investigations. Finally, we here considered active only plaques with TBR ≥ 1. Hitherto, no clear cut-off is known for the detection of SST
2 positive plaques and most evaluation relay on visual interpretation. Previous studies on inflamed plaques considered in the final analysis all lesions irrespective from their TBR [
9,
10]. It should be noted that previous works also considered soft plaques. In this regard, the fact that applying our threshold to calcified plaques yielded significant associations with the degree of calcification and with follow-up data gives reliance in considering it adequate on LAFOV PET to identify conceivably inflamed plaques. The degree of such inflammation is then essential to stratify cardiovascular risk.
While additional prospective head-to-head comparisons with, e.g., [18F]FDG are needed to support our data and implement LAFOV PET/CT in clinical routine, the results of our study support the concept that LAFOV PET systems may serve as an important tool to identify patients at increased risk of MACE.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.