CCR2 Imaging in Human ST-Segment Elevation Myocardial Infarction

Abstract

Among the diverse populations of myeloid cells that reside within the healthy and diseased heart, C-C chemokine receptor 2 (CCR2) is specifically expressed on inflammatory populations of monocytes and macrophages that contribute to the development and progression of heart failure^1–4. Here, we evaluated a peptide-based imaging probe (⁶⁴Cu-DOTA-ECL1i) that specifically recognizes CCR2⁺ monocytes and macrophages for human cardiac imaging. Compared to healthy controls, ⁶⁴Cu-DOTA-ECL1i heart uptake was increased in subjects following acute myocardial infarction, predominately localized within the infarct area, and was associated with impaired myocardial wall motion. These findings establish the feasibility of molecular imaging of CCR2 expression to visualize inflammatory monocytes and macrophages in the injured human heart.

Inflammation contributes to adverse cardiovascular outcomes in patients with atherosclerosis, myocardial infarction (MI), and various forms of heart failure^5,6. Previous studies have established robust associations between serum markers of inflammation (Interleukin-1, Interleukin-6, Tumor Necrosis Factor, C-Reactive Protein) and patient outcomes including major adverse cardiovascular events, heart failure progression, and mortality that were independent of MI severity^6–8. Advanced cardiac imaging has added to these findings to demonstrate the importance of inflammation within the heart. Indeed, inflammation as assessed by magnetic resonance imaging (native T1, T2, late gadolinium enhancement) and positron emission tomography/computed tomography (PET/CT) [(¹⁸F-fluorodeoxyglucose (¹⁸F-FDG)] is associated with accelerated left ventricular (LV) remodeling and reduced LV systolic function following MI^9–13. These observations highlight the potential utility of identifying and suppressing inflammation in the heart.

Until recently, limited information existed regarding the precise immune cell types that are responsible for myocardial inflammation and heart failure progression. Paradigm shifting work has uncovered that the mammalian heart contains abundant and heterogeneous populations of monocytes and macrophages with differing origins, dynamics, and functions. This concept diverges from the previously held view that all macrophages are derived from a single origin and promote inflammation. Under steady-state conditions, the heart contains an admixture of functionally distinct monocyte and macrophage subsets that can be distinguished by the cell surface expression of C-C chemokine receptor 2 (CCR2)^1–4. CCR2⁻ macrophages establish residency within the heart during embryogenesis, are maintained throughout life independent of blood monocyte input, suppress inflammation, and are key regulators of coronary angiogenesis and cardiac tissue repair. In contrast, CCR2⁺ macrophages are derived from monocytes, accumulate in the injured and failing heart, and drive inflammatory responses by generating cytokines (IL-1β, IL-6), oxidative species, and chemokines (CCL2, CCL7, CCL17, CXCL1, CXCL2, CXCL5) that trigger neutrophil and monocyte infiltration, activate fibroblasts, and suppress regulatory T-cell recruitment. Removal of CCR2⁺ macrophages attenuates heart failure through reductions in inflammation, collateral myocardial injury, and fibrosis^2,14,15. These findings have identified CCR2⁺ monocyte and macrophage populations as key drivers of myocardial inflammation. Consequently, there is renewed interest in targeting the immune system in cardiac diseases, a concept that lost momentum following disappointing results from clinical studies investigating corticosteroids, TNF blockade, and cyclosporine in acute MI and heart failure patients^16–18.

Molecular imaging of the heart represents a powerful approach to non-invasively visualize immune cell targets, gain prognostic information, and identify patients best suited to receive immunomodulatory therapies. Among non-invasive imaging approaches, PET is preferred because of its inherent high sensitivity, quantification, and established pathways for clinical translation⁷. Numerous radiotracers including ⁶⁸Ga-pentixafor (CXCR4), ⁶⁸Ga-DOTA-TATE/TOC (somatostatin receptor), ¹¹C-PK11195/¹⁸F-GE180 (mitochondrial translocator protein), ¹⁸F-FDG (glucose metabolism), and ¹¹C-methionine (amino acid metabolism) have been used to detect inflammation⁷. However, they do not identify specific pathological immune populations. Among chemokine receptors, CCR2 is specifically expressed on the surface of inflammatory monocytes and macrophages^1–4. The goal of the current study was to establish the feasibility of visualizing CCR2⁺ monocyte and macrophage recruitment and retention within the human heart following MI.

Results

Recently, we developed a peptide-based probe that specifically binds to CCR2 at an allosteric position (extracellular loop 1) that can be radiolabeled with ⁶⁸Ga or ⁶⁴Cu (⁶⁸Ga/⁶⁴Cu-DOTA-ECL1i)²⁰. ECL1i does not activate CCR2 signaling and instead acts as an inhibitor at high concentration (>5 μM)¹⁹. In vitro cell binding assays in THP-1 cells (express CCR2)²⁰ and 293T cells (do not express CCR2) showed specific binding to CCR2 (IC₅₀: 58.3±18 nM) (Extended data Fig. 1). ⁶⁴Cu-DOTA-ECL1i specifically bound to CD14⁺ monocytes and macrophages and not neutrophils, T-cells, or B-cells in vitro (Extended data Fig. 2). Preclinical studies revealed sensitive and specific detection of CCR2⁺ monocytes and macrophages in rodent heart, atherosclerosis, and lung disease models^21–25. In a Phase 0 trial we reported, ⁶⁴Cu-DOTA-ECL1i recognizes CCR2⁺ monocytes and macrophages in human specimens, is safe in humans, and identified CCR2⁺ cells in patients with interstitial lung disease²⁶.

We enrolled 6 healthy controls and 7 patients that suffered a ST-segment elevation MI (STEMI) (Table 1). Healthy controls subjects were 42–77 years of age without known medical co-morbidities. STEMI patients (35–74 years of age) were enrolled after undergoing percutaneous coronary intervention following STEMI involving a major coronary artery. Individuals with uncontrolled angina, heart failure, hemodynamic or electrical instability, previous STEMI, or those >300 lbs. were excluded. Imaging was performed with ECG-gated ^99mTc-tetrofosmin single photon emission computed tomography/CT (SPECT/CT) to identify the infarct zone, assess its relative regional perfusion and LV function. PET/CT with ⁶⁴Cu-DOTA-ECL1i was performed to identify CCR2 uptake. STEMI patients were imaged across various time points (1–164 days) post-MI. Clinical data was obtained from the medical record.

Table 1.

Clinical demographic table

Subject ID	Age (yrs)	Sex	Race	Weight (kg)	Infarct artery	Medical Comorbidities	Time imaged post-STEMI (days)	Troponin (ng/mL)	WBC (103/ml)	HCT (%)	Platelets (103/ml)	Creatinine (mg/dL)
STEMI-1	74	M	B	88	LAD	HTN, CKD, DM, HLD, Gout	2	29.18	8.8	28.7	182	2.14
STEMI-2	65	M	B	60	LCX	Tobacco	2	85.55	8.2	38.7	203	1.10
STEMI-3	64	M	W	81	LAD	Tobacco, AF	1	30.63	9.7	42.9	211	1.03
STEMI-4	73	M	W	99	LAD	Tobacco, AF	23	267.06	7.5	36.3	265	1.47
STEMI-5	35	M	A	99	LAD	GERD, Hyperglycemia, Anemia	62	272.64	21.8	48.1	235	1.45
STEMI-6	40	M	B	111	LAD	HTN, HF, HLD, DM	164	56.72	9.5	43.1	296	1.08
STEMI-7	56	M	B	87	LCX	COPD, HTN, CKD, DM HLD	101	21.34	14.5	46.8	157	0.96
control-1	77	F	W	78		Healthy volunteer	N/A	N/A	7.1	39.5	205	1.00
control-2	43	M	W	77		Healthy volunteer	N/A	N/A	4.9	44.1	191	0.90
control-3	59	F	W	92		Healthy volunteer	N/A	N/A	5.3	44.7	192	0.80
control-4	56	M	W	77		Healthy volunteer	N/A	N/A	7.1	39.9	184	1.20
control-5	53	M	W	86		Healthy volunteer	N/A	N/A	5.5	44.2	272	1.00
control-6	42	F	W	72		Healthy volunteer	N/A	N/A	7.0	37.3	241	0.70

STEMI: ST-segment elevation myocardial infarction, M: male, F: female, B: black, W: white, A: Asian, LAD: left anterior descending artery, LCX: left circumflex artery, HTN: hypertension, CKD: chronic kidney disease, DM: diabetes mellitus, HLD: hyperlipidemia, AF: atrial fibrillation, GERD: gastroesophageal reflux disease, HF: heart failure, COPD: chronic obstructive pulmonary disease, N/A: not applicable, WBC: white blood cell, HCT: hematocrit. All patients were on statins and low dose aspirin and one control was receiving intermittent low dose oral prednisone.

Healthy controls displayed homogeneous myocardial uptake of ^99mTc-tetrofosmin indicative of normal myocardial perfusion and minimal uptake of ⁶⁴Cu-DOTA-ECL1i consistent with a non-inflammatory state. In contrast, STEMI patients, the infarct zone demonstrated reduced average relative perfusion of 59.5% ±7.4%. Co-registration of SPECT/CT perfusion and ⁶⁴Cu-DOTA-ECL1i PET/CT images suggested that CCR2 tracer uptake was selectively localized to the infarct region and increased relative to the perfusion deficit (Fig. 1A, Extended data Fig. 3–4). Semi-quantitative analysis further supported that STEMI patients exhibited increased ⁶⁴Cu-DOTA-ECL1i uptake in the infarct region compared to healthy controls (SUV_mean 2.53±0.3 vs. 1.36±0.2, p<0.001). Moreover, ⁶⁴Cu-DOTA-ECL1i uptake within the infarct zone was significantly higher than in regions remote from the site of infarction (SUV_mean 2.53±0.3 vs. 1.71±0.5, p=0.001) resulting in an infarct/remote uptake ratio of 1.64±0.7. We did not observe a statistical difference between healthy controls and the remote zone of STEMI patients (p=0.22) (Fig. 1B). Given the exploratory nature of the study, further investigation with a larger sample size is needed to determine if radiotracer uptake in the remote myocardium is higher in STEMI patients compared to controls, a finding that would indicate global inflammation within the heart following myocardial infarction. Consistent with a selective increase in myocardial CCR2 signal, we did not observe a difference in skeletal muscle ⁶⁴Cu-DOTA-ECL1i uptake between controls and STEMI patients (Fig. 1C). ⁶⁴Cu-DOTA-ECL1i uptake in the bone marrow was higher in STEMI patients compared to healthy controls (1.31±0.1 vs. 0.95±0.3, p=0.03). Splenic uptake was similar between controls and STEMI patients (1.78±0.4 vs. 1.72±0.1, p=0.37).

Figure 1. CCR2 imaging in healthy controls and subjects following STEMI.

A, Representative ^99mTc-tetrofosmin (^99mTc) SPECT/CT and differential ⁶⁴Cu-DOTA-ECL1i (CCR2) PET/CT fused images of healthy controls and STEMI patients. SPECT/CT perfusion and CCR2 PET/CT images are co-registered and comparative anatomic slices displayed. Differential ⁶⁴Cu-DOTA-ECL1i images are corrected for blood activity. Green and red arrows denote the infarct region. Color scale bar indicates normalized relative tracer uptake. B, ⁶⁴Cu-DOTA-ECL1i myocardial signal. For STEMI subjects, differential mean standardized uptake value (SUV_mean) was calculated in the infarct and remote regions. Statistical significance was determined using 1-way ANOVA. Control n=6, STEMI n=7. C, ⁶⁴Cu-DOTA-ECL1i skeletal muscle signal. Statistical significance was determined using Mann-Whitney test (2-sided). Control n=6, STEMI n=7. D, ⁶⁴Cu-DOTA-ECL1i myocardial signal as a function of time following STEMI. Each data point represents an individual patient. The mean value for controls with standard deviation is displayed as line with grey zones. STEMI n=7. E, Corresponding 17-segment model polar maps of SPECT/CT perfusion and myocardial wall motion and CCR2 PET/CT from a representative subject. ^99mTc-tetrofosmin SPECT/CT and wall motion polar maps are displayed as % of maximal value (100%). The CCR2 PET/CT polar map represents SUV_mean values. F, Linear regression analysis examining the association between CCR2 tracer uptake and wall motion score. p<0.05 and r²=0.4. n=7. Each data point represents an individual patient. Error bars represent standard deviation.

Intriguingly, ⁶⁴Cu-DOTA-ECL1i uptake within the infarct region was persistently elevated across all examine time points following STEMI (Fig. 1D). The persistent uptake is in contrast to the data determined in mouse models of acute myocardial infarction and suggests that CCR2⁺ monocytes and macrophages may persist in the human heart following myocardial injury²¹. Myocardial segments demonstrating increased ⁶⁴Cu-DOTA-ECL1i uptake had reduced myocardial perfusion and correlated with the degree of regional wall motion abnormalities (r²=0.40, p<0.05) (Fig. 1E–F).

Collectively, our findings establish the feasibility of CCR2 molecular imaging in the human heart. ⁶⁴Cu-DOTA-ECL1i PET/CT performed at various time points following STEMI with percutaneous coronary intervention demonstrated increased differential myocardial tracer uptake compared to controls, showed preferential uptake within the infarcted region, and correlated with impaired myocardial wall motion. These observations are consistent with our pre-clinical studies showing that DOTA-ECL1i radiolabeled with either ⁶⁴Cu or ⁶⁸Ga provides sensitive and specific detection of CCR2⁺ monocytes and macrophages in a wide range of animal models of cardiovascular and pulmonary disease^21–25. Our results build on prior studies indicating that metabolic (¹⁸F-FDG) or C-X-C chemokine receptor type 4 (⁶⁸Ga-pentixafor) imaging is a viable strategy to detect inflammation in the infarct following STEMI^10,22. A limitation of ¹⁸F- FDG and ⁶⁸Ga-pentixafor is that they do not specifically identify pathogenic populations of immune cells. Glucose uptake is also observed in healthy cardiomyocytes, malignancies, infection, and during wound healing^23–25. CXCR4 is expressed on numerous immune cell populations including monocytes, macrophages, neutrophils, T-cells, endothelial cells, and muscle progenitors^11,22. In contrast, ⁶⁴Cu-DOTA-ECL1i recognizes CCR2 which is predominately expressed on a subpopulation of monocytes and macrophages that robustly express inflammatory mediators and contribute to heart failure progression. As such, CCR2 imaging has the potential to identify individuals most likely to respond to immunomodulatory therapies. Lower levels of CCR2 are expressed on basophils and rare populations of T-cells, NK-cells. As the absolute abundance of these subsets within the human heart is substantially lower than CCR2⁺ monocytes and macrophages^26,27, their contribution to CCR2 tracer uptake is relatively minor.

An unexpected finding of our work is that CCR2 uptake persisted for months following the initial ischemic event. While our sample size is small, this observation suggests that CCR2⁺ monocytes and macrophages may persist as a result of either ongoing ischemia or local proliferation²⁸. As such, it is possible that the therapeutic window to deploy immunomodulatory therapies may be larger than previously considered. Consistent with our imaging results, recent single cell RNA sequencing studies revealed expanded numbers of inflammatory populations of monocytes and macrophages within the chronically failing human heart²⁵. CCR2 imaging provides a means to reevaluate the dynamics of myocardial inflammation and relationship with cardiac pathologies including fibrosis in living subjects through serial imaging.

Our study is not without limitations. This was a pilot study to assess the potential of ⁶⁴Cu-DOTA-ECL1i imaging with PET/CT in humans and the sample size was too small to evaluate prognostic implications of CCR2 tracer uptake on LV remodeling, heart failure incidence, and outcomes. Although atherosclerotic plaque inflammation is increased post-STEMI, we did not detect ⁶⁴Cu-DOTA-ECL1i uptake in arterial beds present in the field-of-view. Future studies are needed to address these important questions. While specific, the intensity of ⁶⁴Cu-DOTA-ECL1i signal is not as high as that observed with other radiotracers such as ¹⁸F-FDG and ⁶⁸Ga-pentixafor^9,22, which may impact the sensitivity of detecting CCR2⁺ cell accumulation within the heart. ⁶⁴Cu-DOTA-ECL1i signal within the blood pool may obscure the extent of myocardial uptake measured. In addition, correction of myocardial wall motion may further enhance PET/CT imaging quality. It is possible more quantitative image analytical methods might better detect myocardial ⁶⁴Cu-DOTA-ECL1i signal. However, accurate kinetic modeling requires an arterial input function composed of intact radiotracer, and thus corrected for any radiolabeled metabolites. As we do not have this information, we are unable to perform more complex modeling in these patients. It should be noted the production of blood radiolabeled metabolites will not impact the accuracy of the SUV values.

These limitations non-withstanding, our data establish the feasibility of human CCR2 imaging following STEMI and provide the foundation to explore the natural history and clinical relevance of imaging CCR2⁺ monocytes and macrophages across the spectrum of inflammatory heart diseases.

Methods

The study was approved by the Washington University School of Medicine Institutional Review Board (#201807140) and registered on Clinicaltrials.gov (NCT05107596).

⁶⁴Cu Radiolabeling of DOTA-ECL1i

The ECL1i peptide (LGTFLKC) was synthesized from D-form amino acids by CPC Scientific. DOTA-ECL1i was prepared by conjugating 1,4,7,10-tetraazacyclododecane-1,4,7-tris-acetic acid-10-maleimidoethylacetamide to the cysteine residue of ECL1i following standard operating procedure.²⁹ The crude conjugate was purified by high performance liquid chromatography to reach 99% chemical purity and verified by mass spectrometry. The ⁶⁴Cu radiolabeling of DOTA-ECL1i was performed in ISO class 7 manufacturing suites under current good manufacturing practices (cGMP) conditions within the biological therapy core facility of Siteman Cancer Center by following the batch production record approved by FDA under exploratory investigational new drug application (eIND: 137620). Briefly, 30 μg of DOTA-ECL1i was incubated with 0.925 – 2.59 GBq ⁶⁴CuCl₂ in 500 μL of 20 mM pH 7.0 sodium acetate buffer at 45 °C and shaken at 1000 × g using a thermomixer for 45 min. The chemical identity and radiochemical identity were determined with high performance liquid chromatography equipped with radioactivity and UV detectors and compared to the non-radioactive Cu-DOTA-ECL1i standard. The final product was subject to pre-release quality control analyses including radionuclidic identity, appearance, color, pH, radioactivity strength, filter membrane integrity, radiochemical purity, radiochemical identity, chemical purity, chemical identity, specific activity, and bacterial endotoxin following established standard operating procedures before administration to humans.²⁹ The final product contained 296–370 MBq of ⁶⁴Cu-DOTA-ECL1i with specific activity greater than 18.5 MBq/μg.

In Vitro Cell Binding Assay

The binding affinity of ⁶⁴Cu-DOTA-ECL1i was determined using a CCR2-expressing human monocyte cell line THP-1 while CCR2 negative HEK293T cells (human kidney epithelial cell line) were used as controls²⁰. The half-maximal inhibitory concentration (IC₅₀) values of the ⁶⁴Cu-DOTA-ECL1i were determined using a competitive receptor binding assay with DOTA-ECL1i. Specifically, THP-1 and HEK293T cells (2×10⁶/tube, n=3) in 0.3 mL binding medium (DMEM with 25 mM HEPES, 0.02% BSA, 0.3 mM 1,10-phenanthroline) were incubated with approximately 37 KBq of ⁶⁴Cu-DOTA-ECL1i at room temperature with various concentrations of DOTA-ECL1i ranged from 10⁻⁹ mol/L to 10⁻⁴ mol/L for 1 h. The cells were then centrifuged at 600 × g for 5 min and the supernatants were aspirated. The cell pellets were washed with 400 μL 10 mM PBS (pH7.4) containing 0.2% BSA twice. The bottom of the tubes containing the cell pellets were cut off with a clipper, transferred into plastic tubes, and counted in a well gamma counter (Wallac Wizard 1480, Perkin Elmer). The data were processed with GraphPad Prism to calculate the IC₅₀ values of ⁶⁴Cu-DOTA-ECL1i binding CCR2 (Prism 9.4.0, GraphPad, La Jolla, CA).

Frozen human immune cells isolated from peripheral blood including CD14⁺ cells, T cells, B cells, and neutrophils were purchased from Stemcell Technology (2 donors: male and female, age 65 years old). The cells were thawed first and then transferred into a 50 mL tube with 10 mL of media (PRMI 1640 supplemented with 10% FBS, 100 U/mL penicillin and 100 μg/mL streptomycin, 50 μM 2-mercaptoethanol). The tubes were centrifuged at 1000 × g for 5 min. The cells were reconstructed to a concentration of 2 million/0.2 mL with binding media (DMEM with 25 mM HEPES, 0.02% BSA, 0.3 mM 1,10-phenanthroline). For immune cell binding, 0.2 mL of each type of immune cells were incubated with 37 KBq of ⁶⁴Cu-DOTA-ECL1i for 1 h at room temperature (n=3). For the blocking, the radiotracer was co-added with non-radiolabeled DOTA-ECL1i (1.0 μg) as blockade to 0.2 mL of cells (n=3 technical replicates). After the incubation, the cells were processed and counted as described above.

Subject recruitment

Informed consent was obtained from all subjects prior to enrollment. Seven subjects who had well characterized first time STEMI underwent myocardial ^99mTc-tetrofosmin single-photon emission computed tomography/ computed tomography (SPECT/CT) perfusion and ECG-gated LV functional imaging and ⁶⁴Cu-DOTA-ECL1i PET/CT. All subjects were treated by percutaneous coronary intervention. A single PET/CT was performed in patients at various time points post MI (from 1 day to 164 days). Six healthy control subjects without a known history of cardiovascular disease were recruited as normal controls. Exclusion criteria included prior MI, uncontrolled post-MI angina or heart failure, hypertension (systolic blood pressure >200 mmHg, diastolic blood pressure >110 mmHg, hypotension, (systolic blood pressure <90 mmHg, diastolic blood pressure <50 mmHg), high-grade intermittent AV block without a pacemaker, contraindications to cardiovascular PET/CT imaging such as claustrophobia, recreational drug use, prior STEMI, or a body weight >300lbs (weight limit of PET/CT table).

Imaging Protocol

To localize the site of infarction (patients) and assess LV function, all patients and controls first underwent resting ^99mTc-tetrofosmin SPECT/CT (GE Discovery 670 CZT, GE Healthcare, Waukesha, Wisconsin) myocardial perfusion imaging where 30–45 minutes following the intravenous injection of ^99mTc-tetrofosmin (755–1236 MBq) electrocardiographically gated SPECT/CT images were obtained. Within 24 hours of that study, all subjects underwent a 60-minute dynamic PET/CT scan (Biograph Vision 600, Siemens Healthineers, Knoxville, TN, USA) of the chest following the intravenous administration of 178–355 MBq of ⁶⁴Cu-DOTA-ECL1i. To assess the uptake of the radiotracer in the bone marrow and spleen, a subsequent 20 min PET/CT image was obtained over the abdomen.

Image Analysis

Image analysis was performed by an operator blinded to the designation of control versus STEMI patient. Myocardial ^99mTc-tetrofosmin images were displayed in conventional clinical format with regional activity displayed in reference to 100% of peak activity with the LV was divided in the American Heart Association 17-segment model using standard clinical software. Perfusion defects representing infarction sites were identified by visual inspection and graded on the American Society of Nuclear Cardiology scoring system of 0-normal, 1-mild defect, 3-moderate defect and 4-severe defect³⁰. On the electrocardiographically gated SPECT/CT images systolic function for each segment was visually graded from 1-normal, 2-mild hypokinesis, 3-moderate/severe hypokinesis and 4-akinetic/dyskinetic). Differential myocardial ⁶⁴Cu-DOTA-ECL1i PET images were generated from the scaled subtraction of first 3 minutes (dominated by blood signal) from the last 20 minutes of the data collection. The blood signal in the left ventricle was used for scaling. ⁶⁴Cu-DOTA-ECL1i differential PET images were then reoriented along the standard American Heart Association orientation and divided in the 17-segments model using CARIMAS software (Turku PET Center, https://turkupetcentre.fi/carimas/). Fused differential ⁶⁴Cu-DOTA-ECL1i PET and ^99mTc-tetrofosmin SPECT images were generated using rigid alignment in MIM software (MIM, Cleveland, Ohio). Segments were defined as the site of infarction based on their alignment with the same segments on the perfusion scan. For analysis of PET/CT datasets, standardized uptake values were obtained for infarcted and remote myocardium, LV blood pool as well for the bone marrow and spleen using free-form volumes of interest traced within the boundaries of the organ of interest.

Statistical Analysis

Group variation is described as the mean ± SD. Statistical analysis was performed using GraphPad Prism (version 6.07). The Mann-Whitney U test was used to compare two different groups. One-way ANOVA was used to compare multiple groups. The Pearson product-moment correlation was used to assess the relationship between parameters. The significance level in all tests was a p value of 0.05 or less.

Extended Data

Extended data Figure 1. In vitro cell binding assays of ⁶⁴Cu-DOTA-ECL1i in THP-1 cells and 293T cells.

Data are representative of at least three independent experiments from THP cells (left, express CCR2) and HEK293T cells (right, do not express CCR2). Cells (2 × 10⁶) were incubated with approximately 37 KBq of ⁶⁴Cu-DOTA-ECL1i with the indicated concentrations of cold DOTA-ECL1i at room temperature for 1 h. n=3 technical replicates. Error bars are standard deviation.

Extended data Figure 2. In vitro cell binding assays of ⁶⁴Cu-DOTA-ECL1i in human immune cells.

Data are representative of at least three independent experiments from 2 donors. *** p<0.001, **** p<0.0001. 1-way ANOVA. CPM: counts per minute. n=3 technical replicates. Error bars are standard deviation.

Extended data Figure 3. CCR2 imaging in the other control subjects.

^99mTc-Tetrofosmin (^99mTc) SPECT/CT and differential ⁶⁴Cu-DOTA-ECL1i (CCR2) PET/CT fused images as described in Figure 1. SPECT/CT perfusion and CCR2 PET/CT images are co-registered and comparative anatomic slices displayed. Differential ⁶⁴Cu-DOTA-ECL1i images are corrected for blood activity. Color scale indicates normalized relative tracer uptake.

Extended data Figure 4. CCR2 imaging in the STEMI patients.

^99mTc-Tetrofosmin (^99mTc) SPECT/CT and differential ⁶⁴Cu-DOTA-ECL1i (CCR2) PET/CT fused images as described in Figure 1. SPECT/CT perfusion and CCR2 PET/CT images are co-registered and comparative anatomic slices displayed. Differential ⁶⁴Cu-DOTA-ECL1i images are corrected for blood activity. Green and red arrows denote the infarct region. Color scale indicates normalized relative tracer uptake.

Extended data Table 1.

Imaging tracer activities

Subject ID	99mTc-Tetrofosmin injected dose (MBq)	64Cu-DOTA-ECL1i Injected dose (MBq)
STEMI-1	869.5	355.2
STEMI-2	821.4	336.7
STEMI-3	847.3	296
STEMI-4	754.8	340.4
STEMI-5	869.5	299.7
STEMI-6	1235.8	318.2
STEMI-7	884.3	340.4
control-1	832.5	325.6
control-2	839.9	318.2
control-3	780.7	329.3
control-4	799.2	177.6
control-5	873.2	325.6
control-6	859.5	336.7

Data Availability.

All source data is available from the authors upon request.