Imaging Macrophage and Hematopoietic Progenitor Proliferation in AtherosclerosisNovelty and Significance
Rationale: Local plaque macrophage proliferation and monocyte production in hematopoietic organs promote progression of atherosclerosis. Therefore, noninvasive imaging of proliferation could serve as a biomarker and monitor therapeutic intervention.
Objective: To explore 18F-FLT positron emission tomography–computed tomography imaging of cell proliferation in atherosclerosis.
Methods and Results: 18F-FLT positron emission tomography–computed tomography was performed in mice, rabbits, and humans with atherosclerosis. In apolipoprotein E knock out mice, increased 18F-FLT signal was observed in atherosclerotic lesions, spleen, and bone marrow (standardized uptake values wild-type versus apolipoprotein E knock out mice, 0.05±0.01 versus 0.17±0.01, P<0.05 in aorta; 0.13±0.01 versus 0.28±0.02, P<0.05 in bone marrow; 0.06±0.01 versus 0.22±0.01, P<0.05 in spleen), corroborated by ex vivo scintillation counting and autoradiography. Flow cytometry confirmed significantly higher proliferation of macrophages in aortic lesions and hematopoietic stem and progenitor cells in the spleen and bone marrow in these mice. In addition, 18F-FLT plaque signal correlated with the duration of high cholesterol diet (r2=0.33, P<0.05). Aortic 18F-FLT uptake was reduced when cell proliferation was suppressed with fluorouracil in apolipoprotein E knock out mice (P<0.05). In rabbits, inflamed atherosclerotic vasculature with the highest 18F-fluorodeoxyglucose uptake enriched 18F-FLT. In patients with atherosclerosis, 18F-FLT signal significantly increased in the inflamed carotid artery and in the aorta.
Conclusions: 18F-FLT positron emission tomography imaging may serve as an imaging biomarker for cell proliferation in plaque and hematopoietic activity in individuals with atherosclerosis.
- positron emission tomography
Inflammatory monocytes and macrophages are innate immune cells that promote the growth and complication of atherosclerotic lesions. Once recruited to the arterial wall, mononuclear phagocytes can ingest lipoproteins. Often, the cells produce proinflammatory mediators and differentiate into foam cells. Activated macrophages also elaborate proteases that weaken the plaque’s extracellular matrix.1,2 In early-stage atherosclerosis in mice, most plaque macrophages are direct progeny of recruited blood monocytes that originate in the bone marrow and spleen. In advanced disease, monocyte-derived macrophages proliferate locally, a process that contributes dominantly to the cell population in mature plaques.3 Hence, in early and in late disease stages, cellular proliferation pivotally promotes expansion of both the systemic and local myeloid cell pools. By extension, proliferation likely also drives disease progression. Hematopoietic cell proliferation, either remote in bone marrow and spleen or locally in plaque, may thus represent a therapeutic target.
In This Issue, see p 825
Commonly, we rely on ex vivo cell cycle analysis to measure cellular proliferation. These assays have limited use for assessing proliferative tissue activity in patients with atherosclerosis because the tissues of interest are not readily available for biopsy. With few exceptions, we are currently limited to investigating circulating leukocytes. However, these cells do not proliferate, and monocyte numbers in blood may not faithfully reflect proliferation in plaque or hematopoietic organs. In a hypothetical scenario, plaque macrophage proliferation may be high, despite normal blood monocyte levels, resulting in inflamed vascular lesions that are prone to complications. An imaging biomarker that reports on proliferating cells would expand diagnostic capabilities to disease-relevant tissues and aid development of emerging cardiovascular immuno-modulatory drugs.
Imaging of cancer cells has explored the use of such a biomarker.4 18F-FLT, a positron emission tomography (PET) isotope-labeled thymidine analog, enriches in proliferating cells. Phosphorylation traps the agent intracellularly.5 Thymine is one of the 4 bases needed for DNA replication that occurs before cell division. The small molecule PET agent 18F-FLT has undergone clinical exploration for assessing proliferation, and its retention correlates with the histological cell cycle marker Ki67 in tumor cells, especially in brain, lung, and breast cancer.6 In oncology, 18F-FLT PET imaging has emerged as a promising strategy to monitor treatment response for some cancers that rely more on the DNA salvage pathway than on de novo thymidine synthesis.5,7,8 The comparable proliferative rates of cancer cells and plaque macrophages and previous reports of 18F-FLT uptake into hematopoietic tissues5,9,10 led us to hypothesize that this agent could prove useful for imaging hematopoietic cells in cardiovascular disease. The frequent use of the thymidine analog bromodeoxyuridine (BrdU), an established tool in hematology to measure leukocyte and progenitor proliferation, further suggested that the thymidine analog 18F-FLT may report on macrophage proliferation.
Here we show that atherosclerotic plaques in atherosclerotic apolipoprotein E knock out (ApoE−/−) mice, rabbits, and patients accumulate 18F-FLT. In accordance with the recently reported increase of macrophage proliferation in mature atherosclerotic lesions,3 aortic 18F-FLT retention correlates with the duration of an atherogenic diet in mice and with the Framingham Risk Score in patients. Macrophage-rich plaques with high 18F-FDG signal show the highest 18F-FLT uptake in atherosclerotic rabbit aortae. Hematopoietic organ 18F-FLT uptake associates with hematopoietic activity: it increases in mice with atherosclerosis and after exposure to the Toll-like receptor (TLR) ligand lipopolysaccharide and decreases after injection of fluorouracil (5-FU), an antimetabolite drug that kills proliferating cancer cells and cycling hematopoietic stem and progenitor cells (HSPCs).
A detailed method section is available in the Online Data Supplement.
Female ApoE−/− and wild-type C57BL/6 mice (10–12 weeks old) were purchased from Jackson Laboratories. Atherosclerosis was induced by feeding high-cholesterol diet to ApoE−/− mice ranging from 10 to 28 weeks. In wild-type mice, modulation of hematopoietic progenitor cell proliferation in bone marrow and the spleen was induced by injecting lipopolysaccharide (10 μg/mouse IP on 3 and 1 days before imaging) and 5-FU (150 μg/g IP 12 hours before imaging). To investigate whether the 18F-FLT uptake in atherosclerosis reflects cell proliferation, an additional cohort of ApoE−/− mice was treated with 5-FU at the dose of 150 μg/g IP 12 and 4 hours before injecting 18F-FLT. For cell cycling analysis by flow cytometry, BrdU (DB Biosciences), an antibody-recognizable analog of thymidine, was injected IP in these ApoE−/− mice immediately after 18F-FLT injection. All procedures in mice were approved by the Institutional Animal Care and Use Committee Subcommittee on Research Animal Care, Massachusetts General Hospital, Charlestown, MA.
Atherosclerosis was induced in 4 male New Zealand White rabbits by a combination of high-cholesterol diet and double balloon injury of the abdominal aorta.11 Rabbits were imaged 6 months after diet initiation. Four control rabbits received normal diet. All procedures in rabbits were approved by the IACUC at Mount Sinai Hospital, New York City, NY.
Small Animal PET–Computed Tomography Imaging and Image Analysis
18F-FLT was injected via tail vein 120 minutes before in vivo imaging with an average dose of 22.44±1.07 MBq. This dose and imaging time point were chosen based on published biodistribution data and because mice have high circulating thymidine levels that compete with the PET agent for uptake into proliferating cells.7 Mice were imaged with an Inveon small-animal PET-computed tomography (CT) scanner (Siemens Medical Solutions, Inc, Malvern, PA). CT was performed before PET, acquiring 360 cone beam projections (source power 80 keV and current 500 μA). During CT acquisition, iodine contrast was infused via the tail vein at a rate of 35 μL/min. For quantitative analysis, 1 to 3 regions of interest were drawn manually in the aortic root, the spleen, and the spine of each animal, guided by CT images. After in vivo imaging, mice were euthanized and the direct γ counting was performed on the aortic root, the spleen, and a femur. The data are presented as percent injected dose per gram of tissue (%IDGT).
Animals were subjected to high-resolution 3-dimensional T2-weighted magnetic resonance imaging with a 7-tesla scanner (Magnetom 7T, Siemens Healthcare, Erlangen, Germany). To establish 18F-FLT’s pharmacokinetics, animals were injected with ≈111 MBq, and blood was drawn at 2, 5, 15, 30, 60, 120, and 240 minutes. Radioactivity was measured by a Wizard 2470 Perkin Elmer automatic gamma counter. Values were calculated as %IDGT. All 8 rabbits (average weight 3.47±0.37 kg) underwent 2 PET/CT scans on 2 different days. The first PET/CT was performed after injection of 18F-FDG, whereas the second PET/CT was performed after the injection of 18F-FLT. For both imaging sessions, rabbits were fasted for 4 hours before the isotope injection (water was provided at libitum). Before injection, rabbits were appropriately restrained, and the isotope was injected through a catheter placed in a marginal ear vein. The average injected dose of FDG was 190.3±9.9 MBq (55.2±5.0 MBq/kg), whereas the average injected dose of FLT was 199.2±29.1 MBq (57.5±8.0 MBq/kg). Rabbits were imaged ≈200 minutes after isotope injection (18F-FDG, 197.5±5.50 minutes; 18F-FLT, 198.9±10.5 minutes). Approximately 15 minutes before imaging, rabbits were anesthetized using an injection of ketamine (35 mg/kg) and xylazine (5 mg/kg), and the bladder was emptied. Anesthesia was maintained during imaging using 1% isofluorane breathing anesthesia. PET/CT was performed using a clinical scanner (Siemens Biograph mCT, PET/CT, Siemens, Erlangen, Germany), in 3D mode, using 1 bed position, for 60 minutes. Images were acquired from the thoracic aorta down to the iliac bifurcation. A noncontrast enhanced CT was also performed for attenuation correction of the PET images and to identify the aorta during image analysis.
Human 18F-FDG-PET/CT and 18F-FLT-PET/CT
Twenty patients with paired 18F-FLT-PET/CT and 18F-FDG-PET/CT scans performed between December 2011 and March 2013 were retrospectively included. Ten patients with high risk of cardiovascular disease (atherosclerotic group) and 10 patients with low risk (control group) were enrolled based on their Framingham Score (Table). The study was performed in accordance with the Helsinki Declaration and approved by the regional scientific ethical committee (H-3-2011-092). All patients were diagnosed with neuroendocrine tumors, and written informed consent was obtained from all participants. Patients underwent an 18F-FDG-PET/CT and an 18F-FLT-PET/CT within 2 weeks of each other performed on a Siemens Biograph 40 or 64 PET/CT scanner (Siemens Medical Systems, Erlangen, Germany). Patients were instructed to fast at least 6 hours before tracer injection, and static images were acquired 1 hour postinjection of either 18F-FDG or 18F-FLT. Reconstructed images were analyzed using the OsiriX Lite open-source software (Pixmeo). Image analysis was done in a blinded fashion on anonymized scans. Tracer uptake was quantified as maximum and mean standardized uptake values (SUV max and SUV mean). Regions of interest were drawn on all slices of the ascending aorta and at least 10 consecutive slices of the carotid artery. Averaged SUV max and SUV mean were then calculated for each patient in the 2 target regions based on SUV max and SUV mean values from all regions of interest in the respective region. To correlate the uptake of 18F-FDG and 18F-FLT on a subregional level, matched slices of 18F-FDG and 18F-FLT scans were analyzed side-by-side in the ascending aorta.
Arterial Uptake of 18F-FLT Increases in ApoE−/− Mice
We began by exploring the uptake of 18F-FLT in the thoracic aorta of ApoE−/− mice that consumed a high-fat diet. In these mice, the aortic root harbors plaques inhabited by proliferating macrophages. ApoE−/− but not wild-type mice had areas of increased PET signal that colocalized with the aortic root and the ascending aorta on CT angiography (Figure 1A). The PET-derived SUV measured in the aortic root in ApoE−/− mice significantly exceeded that in wild-type mice (wild-type controls versus ApoE−/− mice, 0.053±0.010 versus 0.169±0.013, P<0.05; Figure 1B). We further observed signal in the thoracic vertebrae and the sternum in both cohorts, a finding that stimulated further exploration. Ex vivo data obtained by scintillation counting (%IDGT, wild-type controls versus ApoE−/− mice, 0.214±0.015 versus 0.376±0.037, P<0.05; Figure 1C) and autoradiography (Figure 1D) confirmed higher activity in the aortae excised from ApoE−/− mice, and peak signal colocalized with Oil Red O stained atherosclerotic lesions. As reported previously, the aortic root contained numerous Ki67+ CD68+ macrophages (Figure 1E).3 Macrophages contributed 73.5±2.8% of all proliferating Ki67+ cells in the aortic wall. We found significantly less proliferating endothelial cells, smooth muscle cells, and T lymphocytes in the atherosclerotic aortic root (Figure 1E and 1F). The fraction of proliferating macrophages was quantified in digested whole aortae by flow cytometric analysis. Significantly more macrophages were in the active S/G2/M phases of the cell cycle in aortae harvested from ApoE−/− mice (Figure 1G and 1H). Previously, we had described that local macrophage proliferation increases in mature atherosclerotic lesions.3 In line with these data, we found a positive correlation between scintillation counts in aortae of mice injected with 18F-FLT and the duration of the atherogenic diet (Figure 1I). In ApoE−/− mice that received 5-FU, which reduces cell proliferation,3 uptake of 18F-FLT in the aortic root was reduced (Figure 2A–2C), in parallel with reduced BrdU+ macrophages detected by flow cytometry (Figure 2D and 2E). Of note, 18F-FLT and BrdU are both thymidine analogs and were injected at the same time. These data suggest that 18F-FLT signal in atherosclerotic mouse aortae mostly reflects local proliferation of macrophages, but other cells, such as endothelial or smooth muscle cells, may also contribute to the PET signal to a smaller degree.
18F-FLT Signal in Hematopoietic Organs of ApoE−/− Mice
The thymidine analog 18F-FLT enriches in organs with high proliferative rates, including the hematopoietic bone marrow.5 Intrigued by our observation of high PET signal in the sternum and vertebrae and because the hematopoietic system supplies monocytes to inflamed atherosclerotic plaque, we systematically compared 18F-FLT uptake in the bone marrow and spleen of ApoE−/− mice to wild-type controls (Figure 3A). In vivo SUV in bone marrow rose in mice with atherosclerosis (wild-type controls versus ApoE−/− mice, 0.132±0.013 versus 0.276±0.020, P<0.05; Figure 3B) as did activity detected by ex vivo scintillation counting (%IDGT of the femur, wild-type controls versus ApoE−/− mice, 0.735±0.053 versus 1.261±0.114, P<0.05; Figure 3C). These observations could result from higher thymidine uptake, DNA synthesis, and proliferation of HSPC. Indeed, flow cytometric analysis paralleled the in vivo PET data because more lineage− Sca-1+ c-Kit+ (LSK) progenitor cells were in the S/G2/M phase of the cell cycle (Figure 3D and 3E). Examination of the spleen yielded a similar pattern: both in vivo PET signal (SUV, wild-type controls versus ApoE−/− mice, 0.059±0.007 versus 0.224±0.013, P<0.05; Figure 3F and 3G), and ex vivo scintillation counting of the organ (%IDGT, wild-type controls versus ApoE−/− mice, 0.455±0.065 versus 1.345±0.134, P<0.05; Figure 3H) revealed higher 18F-FLT uptake. As in the marrow, the higher tracer uptake paralleled increased cycling of LSK in the spleen of ApoE−/− mice (Figure 3I and 3J). These data indicate that 18F-FLT reflects increased hematopoietic activity in the marrow and spleen of mice with atherosclerosis, in accordance with the previous description of increased bone marrow hematopoiesis in atherosclerotic mice12 and the reactivation of extramedullary, that is, splenic hematopoiesis in atherosclerosis.13
Hematopoietic Activation With a TLR Ligand Increases 18F-FLT Signal
Hematopoiesis provides billions of blood cells every day and increases the production of leukocytes according to numerous stimuli. For instance, HSPCs express TLR4 on their surface, and TLR4 ligation increases proliferation, differentiation, and output of mature cells, including monocytes.14 We therefore hypothesized that the changes observed in hematopoietic organs of mice with atherosclerosis could be reproduced with lipopolysaccharide, a bacterial TLR4 ligand that strongly stimulates hematopoiesis.14 Wild-type mice that received lipopolysaccharide showed increased bone marrow uptake of 18F-FLT (Figure 4A and 4B) that paralleled results of ex vivo scintillation counting (Figure 4C). In accordance with the observed uptake of 18F-FLT, a higher fraction of LSK was in the active cell cycle phase after lipopolysaccharide injection (Figure 4D and 4E). TLR4 ligands induce HSPC migration from the bone marrow to the spleen, where the cells seed extramedullary niches. Indeed, splenic uptake of 18F-FLT was increased, as observed by in vivo imaging (Figure 4F and 4G) and ex vivo scintillation counting (Figure 4H). In parallel, the number of cycling LSK increased dramatically in spleens of mice that received lipopolysaccharide (Figure 4I and 4J).
Dampening of Hematopoiesis With 5-FU Decreases 18F-FLT Signal
Next we explored whether 18F-FLT signal might also monitor a decrease of hematopoietic activity. To test this hypothesis, we treated wild-type mice with 5-FU, a clinical drug that is frequently used in hematology research to ablate cycling stem cells. Previously, 5-FU reduced 18F-FLT signal in tumor-bearing mice.15 Treatment with 5-FU reduced bone marrow uptake of 18F-FLT measured in vivo (Figure 5A and 5B) and ex vivo (Figure 5C). As expected, 5-FU decreased proliferation of bone marrow LSK (Figure 5D and 5E). The spleen showed a similar pattern: 5-FU treatment reduced 18F-FLT signal (Figure 5G and 5H) and the frequency of proliferating LSK in parallel (Figure 5I and 5J).
In Vitro 18F-FLT Uptake
To better understand cellular 18F-FLT uptake, we pursued in vitro incubation of the tracer in primary murine cells. Peritoneal macrophages were isolated by negative magnetic bead selection and either treated with 5-FU or not. Scintillation counting revealed that macrophages that were not treated with 5-FU showed higher uptake of the PET tracer (Figure 6A). Fluorescence-activated cell sorter analysis of these samples reported decreased staining for the proliferation marker Ki67 (Figure 6B and 6C). Further, we used positive magnetic bead selection of murine bone marrow to enrich for c-kit+ progenitor cells and Ly6G+ neutrophils. Incubation of these cells with 18F-FLT was followed by scintillation counting. We detected higher uptake of 18F-FLT into c-kit+ cells (Figure 6D), and flow cytometry documented their higher proliferative activity when compared with Ly-G+ cells (Figure 6E and 6F). Of note, these data do not confer specificity of the PET tracer to progenitor cells but rather indicate that proliferating cells, including c-kit+ progenitor cells, have a higher propensity to take up the imaging agent.
18F-FLT PET in Rabbits With Atherosclerosis
To study 18F-FLT plaque uptake in a different species, we subjected 4 rabbits with atherosclerosis and balloon injury and 4 control rabbits without atherosclerosis to PET/CT imaging. Magnetic resonance imaging confirmed that the treatment induced robust atherosclerotic lesions in the infrarenal aorta (Figure 7A). Rabbits first underwent 18F-FDG PET imaging, which showed a heterogeneous pattern of inflammatory activity (Figure 7B). As expected, the 18F-FDG signal was significantly higher in the aortic segments of rabbits with atherosclerosis (SUV mean controls, 0.20±0.002; atherosclerosis cohort, 0.58±0.02, P<0.0001). Immunohistochemistry of antirabbit macrophage (RAM11) staining demonstrated abundant macrophages in the atherosclerosis of these rabbits (Figure 7C). Previous correlation of 18F-FDG to histology in atherosclerotic rabbits16 and in humans17 showed that 18F-FDG PET signal increases with higher macrophage plaque burden. Within 48 hours of the 18F-FDG PET, rabbits were rescanned after injection of 18F-FLT, enabling a comparison of both PET agents. Pharmacokinetic experiments revealed a blood half-life of 20.36±3.54 minutes for 18F-FLT in atherosclerotic rabbits and an optimal injection-imaging sequence of 200 minutes. The aortic 18F-FLT signal was higher in rabbits with atherosclerosis when compared with control rabbits without atherosclerosis (Figure 7D and 7E). We used the 18F-FDG data to classify 3 aortic regions in rabbits with atherosclerosis: low (SUV<0.4), intermediate (0.4–0.6), and high-grade inflammation (SUV>0.6). The vascular territories that were classified as high-grade inflammation showed the highest 18F-FLT signal (Figure 7F and 7G). Linear regression analysis in aortic segments of rabbits with atherosclerosis revealed that 18F-FLT and 18F-FDG signal were weakly positively correlated (Figure 7H), whereas no correlation was detected in control rabbits.
18F-FLT PET in Humans With Atherosclerosis
To assess the potential of 18F-FLT for PET imaging of proliferation in human atherosclerotic plaque, we performed a retrospective analysis of 10 patients with high risk of cardiovascular disease (atherosclerosis group) and 10 patients with a low cardiovascular risk profile (control group) that were originally imaged because of malignant disease and stratified by Framingham Scoring (Table). We found that patients with high cardiovascular risk showed intense calcification, indicative of atherosclerotic lesions (Figure 8A). Patients had undergone both, 18F-FDG and 18F-FLT PET imaging (Figure 8B and 8C). Analysis of the ascending aorta (Figure 8D–8G) and the carotid artery (Figure 8H–8K) showed increased vascular uptake of both PET agents in the individuals with high cardiovascular risk. As observed in rabbits, linear regression analysis in aortic segments showed a weak positive correlation between 18F-FLT and 18F-FDG uptake (Figure 8 L).
Why are we interested in imaging macrophage proliferation? In inflamed tissue, these cells have a short to intermediate life span, ranging from hours to weeks. In mice with atherosclerosis, the entire plaque macrophage population turns over in 1 month.3 Provision of mononuclear phagocytes influences atherosclerosis progression and plaque characteristics.2 Depleting macrophages and monocytes or inhibiting their recruitment to plaque reproducibly diminishes atherosclerosis in animals. Although several interventions may influence macrophage proliferation, noninvasive imaging reporting on this process could markedly facilitate drug development by informing dose selection and providing an early biomarker of effective targeting. Oncology and neurodegenerative research adopted such companion-imaging strategies for rapid, noninvasive feedback on the targeted pathway or molecule in small numbers of patients. Such information would inform study design and the triage of candidates to advance into large-scale trials, enabling a nimbler drug development strategy.18
Tissue-resident macrophages proliferate everywhere.19 All major organ systems, including the vasculature, heart, and brain, host a network of noninflammatory stromal macrophages. These cells have as of yet unclear functions, but they may promote tissue homeostasis, defense, and regeneration.20 We found that most healthy organs had low 18F-FLT baseline uptake, likely reflecting the low proliferative rate of stromal cells and tissue macrophages. The bone marrow, intestine, kidney, and liver furnish notable exceptions, as reported previously.5 These organs have high baseline proliferative activity or participate in the excretion of 18F-FLT.
HSPCs increase proliferative activity in the bone marrow in atherosclerosis after myocardial infarction and after stroke.12,21,22 Circulating danger signals may also directly alert HSPCs to remote organ ischemic or other injury.2 Compromised reverse cholesterol transport induces increased HSPC activity, linking metabolism to monocyte production.12 Monocyte numbers in blood then rise with progressing hyperlipidemia23 and associate with poor prognosis.2 Cardiovascular disease also reactivates splenic leukocyte production. Gradually in mice with atherosclerosis,13 and rapidly after ischemia,22 HSPCs seed the splenic red pulp and expand monocyte production. Splenic HSPC retention relies on vascular cell adhesion molecule-1 expression by CD169+ macrophages.24 GM-CSF, M-CSF, and interleukin-3 mediate splenic myelopoiesis.13,24 Indeed, increased splenic and bone marrow 18F-FDG uptake associates with higher cardiovascular event rates in patients.25 The decisive contribution of local proliferation to the enlarged macrophage pool in atherosclerotic lesions engendered interest in this process.3 In mice with mature lesions, genetic deficiency of scavenger receptor A reduces incorporation of the thymidine analog BrdU into lesional macrophages.3 These pathways may have contributed to the increased 18F-FLT PET signal observed in the present study, and their exploration as therapeutic targets will benefit from an imaging biomarker, such as 18F-FLT PET.
Rabbits with atherosclerosis displayed increased 18F-FLT signal in vascular territories with high 18F-FDG uptake. However, the correlation between both tracers in aortic segments of atherosclerotic rabbits and patients was weak. 18F-FDG is trapped inside cells with high glucose uptake and correlates with macrophage number in the atherosclerotic plaque in rabbits and humans.16,17 Macrophage accumulation depends on increased proliferation and on monocyte recruitment. The observation of matching tracer uptake in some aortic segments supports that 18F-FLT enriches in plaques with high macrophage proliferation because cell growth and proliferation are energy-intensive processes. In distinction to 18F-FLT, 18F-FDG is not specific for proliferating cells. Thus, some aortic segments high in 18F-FDG may have a high glucose uptake but low proliferating rates. Taken together, 18F-FLT may be a useful to study macrophage supply while reducing proliferation of hematopoietic cells. Myocardial uptake of 18F-FDG can interfere with coronary artery imaging, a limitation that does not occur with 18F-FLT.
As indicated by its use for imaging proliferating cancer cells, 18F-FLT is not specific for macrophages or hematopoietic progenitor cells. Parenchymal cells that proliferate in plaque or hematopoietic organs may contribute to the observed signal, including endothelial and smooth muscle cells. This is a significant limitation of 18F-FLT PET imaging, which is perhaps counterbalanced by its clinical availability. Our histological data imply, however, that most Ki67+ plaque cells were macrophages. A strategy to increase specificity for macrophages could rely on multimodal PET/magnetic resonance imaging, in which 18F-FLT PET could be combined with iron oxide macrophage magnetic resonance imaging. 18F-FLT signal colocalization with T2* MR signal decrease may confer increased cellular specificity. 18F-FLT has not shown adverse effects in clinical imaging trials, paving the way for a prospective trial that correlates 18F-FLT signal in plaque and hematopoietic organs with clinical outcome.
Sources of Funding
This work was funded by grants from the National Institute of Health HL114477, HL117829, HL096576, HL118440, HL125703, HL071021, EB009638, the Deutsche Herzstiftung (S/05/12), and the MGH Research Scholar Award.
In August 2015, the average time from submission to first decision for all original research papers submitted to Circulation Research was 14.31 days.
The online-only Data Supplement is available with this article at http://circres.ahajournals.org/lookup/suppl/doi:10.1161/CIRCRESAHA.115.307024/-/DC1.
- Nonstandard Abbreviations and Acronyms
- apolipoprotein E knock out
- hematopoietic stem and progenitor cells
- lineage − Sca-1+ C-kit+ progenitor cells
- positron emission tomography
- standardized uptake value
- Toll-like receptor
- Received June 12, 2015.
- Revision received September 21, 2015.
- Accepted September 22, 2015.
- © 2015 American Heart Association, Inc.
- Swirski FK,
- Nahrendorf M.
- Chalkidou A,
- Landau DB,
- Odell EW,
- Cornelius VR,
- O’Doherty MJ,
- Marsden PK.
- Zhang CC,
- Yan Z,
- Li W,
- et al
- Agool A,
- Schot BW,
- Jager PL,
- Vellenga E.
- Calcagno C,
- Cornily JC,
- Hyafil F,
- Rudd JH,
- Briley-Saebo KC,
- Mani V,
- Goldschlager G,
- Machac J,
- Fuster V,
- Fayad ZA.
- Yvan-Charvet L,
- Pagler T,
- Gautier EL,
- Avagyan S,
- Siry RL,
- Han S,
- Welch CL,
- Wang N,
- Randolph GJ,
- Snoeck HW,
- Tall AR.
- Robbins CS,
- Chudnovskiy A,
- Rauch PJ,
- et al
- Barthel H,
- Cleij MC,
- Collingridge DR,
- Hutchinson OC,
- Osman S,
- He Q,
- Luthra SK,
- Brady F,
- Price PM,
- Aboagye EO.
- Tawakol A,
- Migrino RQ,
- Bashian GG,
- Bedri S,
- Vermylen D,
- Cury RC,
- Yates D,
- LaMuraglia GM,
- Furie K,
- Houser S,
- Gewirtz H,
- Muller JE,
- Brady TJ,
- Fischman AJ.
- Mulder WJ,
- Jaffer FA,
- Fayad ZA,
- Nahrendorf M.
- Nahrendorf M,
- Swirski FK.
- Courties G,
- Herisson F,
- Sager HB,
- Heidt T,
- Ye Y,
- Wei Y,
- Sun Y,
- Severe N,
- Dutta P,
- Scharff J,
- Scadden DT,
- Weissleder R,
- Swirski FK,
- Moskowitz MA,
- Nahrendorf M.
- Dutta P,
- Hoyer FF,
- Grigoryeva LS,
- et al
Novelty and Significance
What Is Known?
Macrophages promote disease progression in atherosclerosis.
The turnover of macrophages is rapid, making it interesting to study their supply.
Macrophages derive either from monocytes, which are progeny of hematopoietic stem cells in the marrow and spleen, or, especially in mature plaque, from local proliferation.
What New Information Does This Article Contribute?
18F-FLT, a thymidine analog that avidly incorporates into proliferating cells, enriches in atherosclerotic plaque of mice, rabbits, and human patients.
Although other stromal cells contribute, the majority of proliferating cells in atherosclerotic lesions are macrophages.
18F-FLT positron emission tomography (PET)-CT imaging reports increased activity in the hematopoietic organs of mice with atherosclerosis.
18F-FLT plaque uptake increases with the duration of diet exposure in apolipoprotein E knock out mice.
Originally developed for imaging proliferation in malignant disease, this PET imaging method may allow monitoring the expansion of hematopoietic cells in the setting of atherogenesis.
Myeloid cells are key protagonists in the progression of atherosclerotic disease. In inflammation, their life span is considerably shortened, which motivated the field’s recently emerging interest in the production of these cells. The small molecule positron emission tomography tracer 18F-FLT is an analog to the DNA building block thymidine. Just like bromodeoxyuridine, another thymidine analog used to measure cell proliferation, it incorporates into cells that are actively cycling. Our data suggest that 18F-FLT positron emission tomography-CT imaging may be a clinically available method to study hematopoietic activity in patients with atherosclerosis. This could be of particular value in testing novel antiatherosclerotic therapies designed to diminish macrophage infiltration into inflammatory atherosclerotic plaques.