Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Radiosynthesis and preclinical evaluation of [68Ga]Ga-NOTA-folate for PET imaging of folate receptor β-positive macrophages


Folate receptor β (FR-β), a marker expressed on macrophages, is a promising target for imaging of inflammation. Here, we report the radiosynthesis and preclinical evaluation of [68Ga]Ga-NOTA-folate (68Ga-FOL). After determining the affinity of 68Ga-FOL using cells expressing FR-β, we studied atherosclerotic mice with 68Ga-FOL and 18F-FDG PET/CT. In addition, we studied tracer distribution and co-localization with macrophages in aorta cryosections using autoradiography, histology, and immunostaining. The specificity of 68Ga-FOL was assessed in a blocking study with folate glucosamine. As a final step, human radiation doses were extrapolated from rat PET data. We were able to produce 68Ga-FOL with high radiochemical purity and moderate molar activity. Cell binding studies revealed that 68Ga-FOL had 5.1 nM affinity for FR-β. Myocardial uptake of 68Ga-FOL was 20-fold lower than that of 18F-FDG. Autoradiography and immunohistochemistry of the aorta revealed that 68Ga-FOL radioactivity co-localized with Mac-3–positive macrophage-rich atherosclerotic plaques. The plaque-to-healthy vessel wall ratio of 68Ga-FOL was significantly higher than that of 18F-FDG. Blocking studies verified that 68Ga-FOL was specific for FR. Based on estimations from rat data, the human effective dose was 0.0105 mSv/MBq. Together, these findings show that 68Ga-FOL represents a promising new FR-β–targeted tracer for imaging macrophage-associated inflammation.


Over-expression of folate receptor (FR) on cancer cells and during inflammatory responses has been used as a diagnostic and therapeutic tool to enable targeted delivery to tumors and sites of inflammation1. The beta isoform of the folate receptor (FR-β), distinctly expressed on activated macrophages, is a promising imaging marker for inflammatory conditions such as rheumatoid arthritis2. Currently, imaging of inflammation by positron emission tomography/computed tomography (PET/CT) is mainly performed using the glucose analog 2-deoxy-2-[18F]fluoro-D-glucose (18F-FDG), which reflects high consumption of glucose by macrophages and other inflammatory cell types. However, due to its non-specific nature, it is difficult to use 18F-FDG to detect inflammation adjacent to metabolically active tissues such as the heart. Therefore, the development of PET tracers targeting exclusive markers is vital for specific detection of inflammation. The FR-β–targeted PET tracers investigated for this purpose so far include [18F]AlF-NOTA-folate (18F-FOL)3,4,5, [18F]fluoro-PEG-folate6, and 3′-aza-2′-[18F]-fluoro-folic acid (18F-AzaFol)7,8; the latter two have already reached the initial clinical phase9,10. Other recently developed FR-targeted tracers include reduced 18F-folate conjugates11, 55Co-labeled albumin-binding folate derivatives12, and [68Ga]NOTA-folate13, all of which have been investigated in preclinical studies for imaging of FR-overexpressing tumors. Previously, we successfully used 18F-FOL PET to visualize FR-β–positive macrophages in mouse and rabbit models of atherosclerosis3. Atherosclerotic lesions exhibit chronic inflammation associated with accumulation of macrophages in the affected area, providing a rationale for investigating macrophage-targeted tracers.

In radiosynthesis of 68Ga-radiopharmaceuticals, 68Ge/68Ga-generators are commonly used to obtain 68Ga-radionuclides; importantly, these generators can be conveniently implemented in a laboratory setting. In 68Ge/68Ga-generators, a certain amount of 68Ge is immobilized on a stationary phase, where the mother radionuclide decays into 68Ga, which can be eluted out for radiolabeling reactions. In this study, we prepared [68Ga]Ga-NOTA-folate (68Ga-FOL, Fig. 1) with generator-produced 68Ga and evaluated its potential for imaging of inflammation. 68Ga-FOL shares the same precursor structure as 18F-FOL3. Al18F labeling of NOTA-conjugates requires cyclotron facilities for the production of [18F]fluorine, whereas generator-produced 68Ga offers a convenient and cost-effective option for radiolabeling. First, we determined the binding affinity of 68Ga-FOL to human FR-β using transfected cells. Next, we investigated the uptake and specificity of intravenously (i.v.) administered 68Ga-FOL for the detection of inflamed atherosclerotic lesions in mice and compared the tracer with 18F-FDG. In addition, we determined the whole-body distribution kinetics in healthy rats, with and without the blocking agent folate glucosamine, and estimated the human radiation dose of 68Ga-FOL.

Figure 1

(A) The chemical structure of NOTA-Folate. The molecular weight of 68Ga-FOL is 908.82 Da. (B) Radio-HPLC chromatogram of 68Ga-FOL and (C) UV-chromatogram (280 nm) of 1 nmol NOTA-Folate precursor.

Experimental section

General materials and equipment

NOTA-folate precursor was synthesized as previously described14. 68GaCl3 was obtained from 68Ge/68Ga IGG-100 generators (Eckert & Ziegler, Valencia, CA, USA) by elution with 0.1 M hydrochloric acid (HCl) in water. TraceSELECT water (Honeywell, Morristown, NJ, USA) was used for radiosynthesis. Other chemicals were purchased from commercial suppliers. Chinese hamster ovary (CHO) cells stably transfected with human FR-β (CHO-hFRb; CHO-FR-β+) were a generous gift from Philip S. Low, Purdue University, USA. FR-β–negative CHO cells (CHO-FR-β control) were a generous gift from Sirpa Jalkanen, MediCity Research Laboratory, University of Turku, Finland. A dedicated small animal PET/CT (Inveon Multimodality; Siemens Medical Solutions, Knoxville, TN, USA) was used for PET/CT imaging, and a gamma counter (1480 Wizard 3″, PerkinElmer/Wallac, Turku, Finland or Triathler 3″, Hidex, Turku, Finland) was used for radioactivity measurement of ex vivo tissues, blood, and plasma samples. Tracer quality control and plasma metabolite analysis were performed using a LaChrom high-performance liquid chromatography (HPLC) system (Hitachi; Merck, Darmstadt, Germany) equipped with a Radiomatic 150TR flow-through radioisotope detector (Packard, Meriden, CT, USA) (radio-HPLC). Photomicroscopy images were taken with a digital slide scanner (Pannoramic 250 Flash or Pannoramic P1000; 3DHistec Ltd., Budapest, Hungary).

68Ga-FOL radiosynthesis

A fraction of 68Ga-eluate (0.5–1.0 mL) was mixed with an aqueous solution of 2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid (HEPES, 1.2 g/mL in 50–100 µL). NOTA-folate precursor (10–20 nmol in 20–40 µL water) was added, vortexed, and incubated for 10 min at 80 °C. The mixture was then cooled down and brought to a pH of ~ 6.5 by addition of 55 µL of 1 M sodium hydroxide (NaOH). The product was used without further purification. Radiochemical purity was analyzed primarily by HPLC. The HPLC conditions were as follows: 250 × 4.6 mm Jupiter Proteo 4µ C18 90 Å column (Phenomenex, Torrance, CA, USA); flow rate = 1 mL/min; wavelength λ = 220 nm; solvent A = 0.1% trifluoroacetic acid (TFA) in water; solvent B = 0.1% TFA in acetonitrile; gradient: during 0–14 min from 3% B to 35% B; during 14–15 min from 35% B to 3% B. Representative radio- and UV chromatograms are presented in Fig. 1B,C, respectively.

During radiosynthesis set-up, the radiochemical purity was also analyzed by instant thin-layer chromatography (iTLC) (Supplementary Figure S1). A 1.0 µL sample of the end product or reaction mixture was applied to a silica gel-based iTLC strip (iTLC-SG; Agilent, Santa Clara, CA, USA) and developed with 50 mM citric acid. Unbound 68Ga migrated up with the mobile phase with a retention factor (Rf) of 0.8–1.0, while 68Ga-FOL remained at the application point (Rf = 0). To measure the unbound and tracer-bound 68Ga fractions, the strip was cut into two pieces along a line midway between the baseline and the solvent front, and each piece was measured separately in a gamma counter.

68Ge breakthrough was monitored by collecting aliquots from the 68Ge/68Ga generator eluate and measuring their radioactivities at time of collection and again 24 to 48 h later.

The lipophilicity of 68Ga-FOL (distribution coefficient LogD) was determined as previously described15. To evaluate 68Ga-FOL stability in the injectable formulation, we kept the end product at room temperature (RT) and took samples for radio-HPLC analysis at time intervals of up to 3 h.

Quantification of Affinity of 68Ga-FOL for FR-β

The binding specificity of 68Ga-FOL to FR-β was evaluated using CHO-FR-β+ and CHO-FR-β cells (control). The cells were cultured at 37 °C in a CO2 incubator in RPMI 1640 medium (Gibco/Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS; Biowest, Nuaillé, France).

To verify FR-β expression, the cultured cells were harvested and incubated with either fluorescein isothiocyanate (FITC)-conjugated anti-human FR-β antibody (m90916, a gift from Philip S. Low) or allophycocyanin (APC)-conjugated anti-human FR-β antibody (mouse IgG2a; BioLegend, San Diego, CA, USA) or the corresponding isotype controls (mouse IgG-FITC, mouse IgG2a-APC; BioLegend). The cells were fixed using paraformaldehyde and analyzed using a Fortessa fluorescence-activated cell sorting (FACS) device (BD Biosciences, Franklin Lakes, NJ, USA) and the Flowing software (Cell Imaging and Cytometry Core, Turku Bioscience, Turku, Finland).

After verifying the presence of FR-β on the cells, either CHO-FR-β+ or CHO-FR-β cells were cultured on one side of a 92 mm petri dish in a tilted position in growth medium at 37 °C in a CO2 incubator. The other side of the petri dish with no cells was used as background control for non-specific binding of 68Ga-FOL. Once the cells grew to a confluent monolayer, the growth medium from the petri dishes was removed, and phosphate-buffered saline (PBS) containing calcium and magnesium with 10% FBS (binding medium) was added. To starve the cells of folate, the dishes were incubated for 30 min at 37 °C in a CO2 incubator. After incubation, the cells were rinsed with binding medium (2 × 2 mL). A LigandTracer Yellow instrument (Ridgeview Instruments AB, Uppsala, Sweden) was used to measure the dissociation constant (KD) for 68Ga-FOL. The assay protocol with LigandTracer Yellow involves consecutive radioactivity measurements of the target (cell region) and the background (i.e., the cell-free region on the petri dish). Radioactivity was measured in each region for 30 s as raw counts per second (cps) with a delay of 5 s over the time course of the experiment. Target regions (cps) were corrected for background signal and radioactive decay. To detect background radioactivity or noise picked up by the instrument, 5 mL binding medium was added to the cells on the petri dish. After 15 min, 68Ga-FOL was added stepwise to achieve a concentration range of 1 to 80 nM, followed by replacement with fresh binding medium to measure the dissociation. The ratio of bound 68Ga-FOL (to the cells) to background (petri dish) and the KD value were calculated using the TraceDrawer software (Ridgeview Instruments AB).

Animal experiments

Low-density lipoprotein receptor-deficient mice expressing only apolipoprotein B100 (LDLR−/−ApoB100/100, strain #003000; Jackson Laboratory, Bar Harbor, ME, USA) were used to induce atherosclerosis. The mice were fed a high-fat diet (HFD; 0.2% total cholesterol, TD 88137, Envigo, Madison, WI, USA) starting at the age of 2 months and maintained for 3–5 months. C57BL/6JRj mice (Central Animal Laboratory of the University of Turku) fed with a regular chow diet were used as healthy controls. In total, 17 LDLR–/–ApoB100/100 (34.7 ± 5.5 g) and six healthy control mice (29.65 ± 1.9 g) were studied. In addition, six Sprague–Dawley rats (135.9 ± 17.1 g) from the Central Animal Laboratory of the University of Turku were studied.

All animals were housed at the Central Animal Laboratory of the University of Turku, and had ad libitum access to water and food throughout the study. All animal experiments were approved by the national Animal Experiment Board in Finland (license number ESAVI/4567/2018) and were carried out in compliance with European Union Directive 2010/63/EU.

Mouse studies

PET/CT Imaging

The mice were fasted for 4 h prior to imaging, anesthetized with isoflurane (4–5% induction, 1–2% maintenance), and placed on a heating pad. The mice then received i.v. 18F-FDG (14.4 ± 0.2 MBq) via a tail vein cannula; the next day under the same conditions, they received 68Ga-FOL (20.1 ± 1.0 MBq). Immediately after PET, an iodinated contrast agent (100 µL eXIATM160XL; Binitio Biomedical, Ottawa, ON, Canada) was i.v. injected, and high-resolution CT was performed for anatomical reference. The Carimas 2.10 software (Turku PET Centre, Turku, Finland, was used to analyze PET/CT images. We defined regions of interest (ROIs) for the myocardium in coronal PET/CT images using the contrast-enhanced CT as an anatomical reference, as previously described2. The results were normalized against the injected radioactivity dose and animal body weight, i.e., the data were expressed as standardized uptake values (SUVs).

Ex vivo biodistribution

To study the specificity of 68Ga-FOL uptake, an in vivo blocking study was performed with another group of HFD-fed LDLR–/–ApoB100/100 mice i.v. injected with 68Ga-FOL alone or 68Ga-FOL in conjunction with a 100-fold molar excess of folate glucosamine. Mice were i.v. injected with 68Ga-FOL (11.3 ± 0.8 MBq) and euthanized after 60 min. Various tissues were excised and weighed, and their radioactivity was measured with a γ-counter (Triathler 3″, Hidex). After compensating for radioactivity remaining in the tail and cannula, the ex vivo biodistribution of 68Ga-radioactivity results were expressed as SUVs, and blocking and non-blocking results were compared.

Autoradiography, histology, and immunostainings

Following PET/CT imaging, the dissected aortic arch was processed into 20 and 8 µm cryosections. The 20 µm cryosections were used for digital autoradiography analysis as previously described3. Briefly, the sections were apposed on an Imaging Plate BAS-TR2025 (Fuji, Tokyo, Japan), and the plates were subsequently scanned on Fuji Analyzer BAS-5000 after an exposure time of 3 h for 68Ga-FOL and at least 4 h for 18F-FDG. After scanning, sections were stored at –70 °C until staining with hematoxylin–eosin (H&E) and then scanned with a Pannoramic digital slide scanner. Autoradiographs were analyzed using the Tina 2.1 software (Raytest Isotopemessgeräte, GmbH, Straubenhardt, Germany), and the uptake of 68Ga-FOL and 18F-FDG was corrected for injected radioactivity dose per unit body mass and radioactive decay during exposure; data were expressed as photostimulated luminescence per square millimeter (PSL/mm2). For immunohistochemistry, adjacent 8 µm sections were used to investigate co-localization of 68Ga-FOL with Mac-3-positive macrophages. The sections were incubated with anti-mouse Mac-3 antibody (1:1,000; BD Biosciences, Franklin Lakes, NJ, USA), and a color reaction was subsequently developed using 3.3′-diaminobenzidine (Bright-DAB, BS04-110).

In vivo stability

To determine the in vivo stability of 68Ga-FOL, plasma samples collected from atherosclerotic mice (n = 3) at 60 min post-injection were analyzed using radio-HPLC. Blood samples were collected in heparinized tubes and centrifuged at 4 °C for 5 min at 2,118 × g. Plasma proteins were precipitated with 10% sulfosalicylic acid (1:1 v/v), followed by centrifugation at RT for 2 min at 14,000×g. The supernatant was analyzed by radio-HPLC. Standard samples were prepared by addition of 68Ga-FOL tracer to 500 µL plasma supernatant collected from mice that had not received tracer. Both standard and metabolite samples applied to radio-HPLC analysis were normalized to a final volume of 1 mL by dilution with radio-HPLC solvent A if necessary. The radio-HPLC conditions were as follows: 250 × 10 mm Jupiter Proteo 5µ C18 90-Å column (Phenomenex); flow rate = 5 mL/min; solvent A = 0.1% TFA in water; solvent B = 0.1% TFA in acetonitrile; gradient, 0–11 min from 3% B to 25% B, 11–12 min from 25% B to 100% B, 12–14 min 100% B.

Rat studies

To determine distribution kinetics and estimate human radiation dose, dynamic whole-body 68Ga-FOL PET/CT was performed in six healthy rats. In addition, three of the rats were also subjected to a blocking experiment with co-injection of a 100-fold excess of folate glucosamine. The rats were injected with 10.3 ± 0.4 MBq of 68Ga-FOL and PET imaged for 60 min. After imaging, the rats were euthanized, and various tissues were excised, weighed, and measured for radioactivity. Plasma samples were analyzed by radio-HPLC as described above. Using CT as an anatomical reference, quantitative PET image analysis was performed by defining ROIs on the main organs, and time–activity curves were extracted with the Carimas software. Human radiation dosimetry was estimated from the rat data using the OLINDA/EXM 2.2 software17.

Statistical analysis

Results are presented as means ± SD. Differences between groups were analyzed by the unpaired Student t test using Microsoft Excel. P values < 0.05 were considered statistically significant.



68Ga-FOL was produced with a radioactivity concentration of 345.8 ± 118.9 MBq/mL (n = 13) and high radiochemical purity (96.6% ± 2.7%). Molar activity was 21.8 ± 6.9 GBq/µmol at the end of radiosynthesis. 68Ge-breakthough remained under 0.001% throughout the study. The distribution coefficient (Log D) of 68Ga-FOL was − 3.28 ± 0.33 (n = 3), indicating high hydrophilicity.

In vitro quantification of 68Ga-FOL binding affinity to FR-β

FACS analyses confirmed that FR-β was clearly expressed on CHO-FR-β+ cells but not on CHO-FR-β cells (Fig. 2A–C). In binding assays, with a stepwise increase in the concentration of 68Ga-FOL from 1 to 80 nM, binding of 68Ga-FOL to CHO-FR-β+ cells was gradually increased and exhibited a KD of 5.1 ± 1.1 nM (n = 3). By contrast, we observed no clear accumulation of 68Ga-FOL on CHO-FR-β cells, even at concentrations up to 40 nM (Fig. 2D).

Figure 2

(A) Representative flow cytometry data showing FR-β expression on human FR-β–negative and FR-β–positive CHO cells stained with FITC-conjugated (red) or APC-conjugated anti–FR-β antibody (blue) or the corresponding isotope controls (black). Quantification of FR-β expression is presented as (B) percentage and (C) mean fluorescence intensity. (D) Representative real-time binding affinity of 68Ga-FOL, measured using LigandTracer. The graph generated with TraceDrawer shows raw counts per second (cps) after correction for background signal and radioactive decay.

68Ga-FOL detects macrophage-rich lesions in atherosclerotic mice

We evaluated the biodistribution of i.v.-administered 68Ga-FOL in mice using in vivo PET/CT, ex vivo gamma counting of excised tissues, and ex vivo autoradiography of aorta cryosections. To study the specificity of 68Ga-FOL to FR-β, we blocked by co-injecting a molar excess of folate glucosamine. In addition, we compared 68Ga-FOL with 18F-FDG in a head-to-head PET/CT imaging setting, as well as by ex vivo autoradiography.

The in vivo stability of 68Ga-FOL was high: 60 min after i.v. injection, the amount of intact tracer was 63.3% ± 1.2% of total plasma radioactivity in LDLR–/–ApoB100/100 mice (n = 3, Supplementary Figure S2).

Our ex vivo results revealed that the aortic uptake of 68Ga-FOL was higher in atherosclerotic mice (SUV 0.75 ± 0.12) than in healthy controls (SUV 0.41 ± 0.10, P = 0.004) or atherosclerotic mice from the blocking study (SUV 0.09 ± 0.03, P = 0.001). Furthermore, the concentration of radioactivity was threefold greater in atherosclerotic aorta than in blood (SUV 0.23 ± 0.09). The highest radioactivity uptake was in FR-positive kidneys18 in both atherosclerotic and control mice (SUV 22.30 ± 3.28 and 20.27 ± 5.48, respectively, P = 0.49), and uptake was significantly reduced in a blocking study performed in atherosclerotic mice (SUV 2.65 ± 1.80, P = 0.0002). The radioactivity of other tissues was much lower than that of the kidneys. Besides the kidneys, blocking with folate glucosamine in atherosclerotic mice decreased the radioactivity concentration in many other tissues as well, including an 88% reduction in the aorta (Supplementary Table S1).

A comparison of the two tracers by in vivo PET/CT revealed that myocardial uptake of 68Ga-FOL (SUV 0.43 ± 0.06) was significantly lower than that of 18F-FDG (SUV 10.6 ± 1.88, P = 0.001, Fig. 3A,B).

Figure 3

(A) Coronal PET/CT and CT images of an atherosclerotic mouse that was administered 68Ga-FOL or 18F-FDG. White arrows show the aortic arch, and red arrows show the myocardial region. (B) Quantification of myocardial PET data showing a significant difference between the tracers. (C) Hematoxylin–eosin (H&E) staining and autoradiography images from representative aorta cryosections, and Mac-3 macrophage marker staining in consecutive aorta cryosections. Black rectangles in the images indicate the plaque region, which are shown at higher magnification. Scale bar = 0.5 mm. A = arch; AA = ascending aorta; B = brachiocephalic artery; D = descending thoracic aorta; L = lesion; LC = left common carotid artery; LS = left subclavian artery; W = wall. (D) Quantification of autoradiography data showing a significant difference between the tracers.

To further elucidate 68Ga-FOL and 18F-FDG uptake in the aortas of atherosclerotic mice in greater detail, we analyzed radioactivity using autoradiography and H&E staining of aortic cryosections followed by macrophage-detecting immunohistochemistry on adjacent tissue cryosections. The results revealed that 68Ga-FOL and 18F-FDG radioactivity co-localized with Mac-3–positive macrophage-rich plaques (Fig. 3C). The plaque-to-healthy vessel wall ratio of 68Ga-FOL (2.44 ± 0.15) was significantly higher than that of 18F-FDG (1.93 ± 0.22, P = 0.005, Fig. 3D).

Distribution kinetics in rats and estimation of the human radiation dose of 68Ga-FOL

In rats as in mice, 68Ga-FOL underwent fast renal excretion and highest uptake in kidneys, urine, salivary glands, liver, and spleen (Fig. 4). Co-injection of 68Ga-FOL along with a molar excess of folate glucosamine clearly decreased tracer uptake in several organs but increased urinary excretion. Radio-HPLC analysis of plasma samples revealed that 68Ga-FOL was relatively stable in vivo (Supplementary Figure S2); at 60 min post-injection, 71.8% ± 1.5% of the total radioactivity was from the intact tracer in healthy rats (n = 3) without blocking, and 88.0% ± 0.7% (n = 3, P = 0.0002) when blocked with folate glucosamine. Extrapolating from the rat PET data, the estimated human effective dose for a 73 kg man was 0.0105 mSv/MBq. The most critical organ was the kidney (0.1420 mSv/MBq) (Supplementary Table S2).

Figure 4

Time–activity curves of healthy rat tissues after i.v. injection of 68Ga-FOL, or co-injection of 68Ga-FOL and molar excess of folate glucosamine (blocking experiment).


In this study, 68Ga-FOL was conveniently prepared with 68Ga-eluate from a 68Ge/68Ga-generator based on a fractionation elution method. We observed that 68Ga-FOL binds to FR-β with high affinity and accumulates in atherosclerotic lesions in mice following i.v. administration. Importantly, 68Ga-FOL exhibited lower myocardial uptake and higher plaque-to-healthy vessel wall ratio than 18F-FDG. Based on estimation from the rat data, the human radiation dose of 68Ga-FOL was low.

To produce 68Ga-FOL, we used a fractionation method to obtain 68GaCl3 from a 68Ge/68Ga-generator for a chelation reaction with the precursor compound NOTA-folate. This is a well-established method that we used previously, and the total radiosynthesis takes less than 20 min.

Previously, two folate-based imaging agents for single-photon emission computed tomography (SPECT), 99mTc-EC20 and 111In-EC0800, were used to detect atherosclerotic lesions in mice19,20. Additionally, the PET tracer 3′-aza-2′-18F-fluorofolic acid can detect FR-β–positive macrophages in human atherosclerotic plaques in vitro8. In our previous studies, we demonstrated that 18F-FOL is specific for FR-β–positive macrophages and can detect inflamed atherosclerotic plaques in mice and rabbits, as well as in human tissue Sects. 3. However, the affinity of 68Ga-FOL binding to human FR-β has not been previously evaluated, and its ability to detect atherosclerotic lesions has not been compared with that of 18F-FDG. Our in vitro binding assay of 68Ga-FOL with CHO-FR-β+ and CHO-FR-β cells revealed specificity and high affinity for FR-β (5.1 ± 1.1 nM), close to the binding affinity of 18F-FOL (1.0 nM) for FR-positive tumor xenografts reported earlier14. Blocking studies in mice and rats further supported the tracer’s specificity for FR, and mouse studies confirmed the ability to detect macrophage-rich inflammatory lesions. The observed low myocardial uptake is beneficial for detection of atherosclerotic lesions in coronary arteries in prospective PET/CT studies of patients with coronary heart disease. When compared with our earlier 18F-FOL study, 68Ga-FOL exhibited similar plaque-to-healthy vessel wall ratios (2.44 ± 0.15 for 68Ga-FOL and 2.60 ± 0.58 for 18F-FOL). However, the in vivo stability of 68Ga-FOL in mice (63% ± 1% intact tracer at 60 min post-injection) was slightly lower than in our previous studies with 18F-FOL (85% ± 6% at 60 min post-injection)3. The human effective dose of 68Ga-FOL extrapolated from the rat data (0.0105 mSv/MBq) is low and within the same range as other 68Ga-tracers21,22,23.

We have obtained 68Ga-FOL with moderately high molar activity but the impact of molar activity on the imaging performance was not evaluated. On the other hand, we are also aware that high > 1,000 GBq/µmol molar activity does not always improve in vivo imaging, as reported by Wurzer and co-workers24. Our tracer, 68Ga-FOL, has directly conjugated NOTA, i.e. one carboxylic arm is used to form an amide bond, which makes the structure less optimal for 68Ga chelation than the N3O3 hexadentate coordination25. We did not explore if a higher molar activity could have been achieved by replacing the NOTA with for example a 1,4,7-triazacyclononane,1-glutaric acid-4,7-acetic acid (NODAGA)26 or triazacyclononane-phosphinate (TRAP)27 chelator, which conserve the N3O3 coordination for 68Ga binding. This is our first-generation conjugate and further optimizations with better chelators such as NODAGA, or even more recent ones such as tris(hydroxypyridinone) (THP) and desferrioxamine (DFO)28 are indeed warranted. In this particular lead compound, NOTA was chosen since it is a commonly used chelator for radiolabeling with both [18F]AlF and 68Ga. Although NOTA is not the most optimized chelator for 68Ga, it is indeed applicable in the preparation of 68Ga-radiopharmaceuticals for clinical use29.


In summary, we have prepared 68Ga-FOL and evaluated its FR-β targeting and imaging performance in vitro and in vivo. The preclinical results of 68Ga-FOL are in line with those of our previous studies using 18F-FOL and corroborate FR-β as an imaging target for detection of inflamed atherosclerotic lesions.


  1. 1.

    Low, P. S. & Kularatne, S. A. Folate-targeted therapeutic and imaging agents for cancer. Curr. Opin. Chem. Biol.13, 256–262 (2009).

    Article  CAS  Google Scholar 

  2. 2.

    Chandrupatla, D. M. S. H., Molthoff, C. F. M., Lammertsma, A. A. & van der Laken, C. J. The folate receptor β as a macrophage-mediated imaging and therapeutic target in rheumatoid arthritis. Drug Deliv. Transl. Res9, 366–378 (2019).

    Article  CAS  Google Scholar 

  3. 3.

    Silvola, J. M. U. et al. Aluminum fluoride-18 labeled folate enables in vivo detection of atherosclerotic plaque inflammation by positron emission tomography. Sci. Rep.8, 9270 (2018).

    Article  CAS  Google Scholar 

  4. 4.

    Elo, P. et al. Folate receptor-targeted positron emission tomography of experimental autoimmune encephalomyelitis in rats. J. Neuroinflammation16, 1–18 (2019).

    Article  CAS  Google Scholar 

  5. 5.

    Jahandideh, A. et al. Folate receptor β targeted PET imaging of macrophages in autoimmune myocarditis. J. Nucl. Med. (2020).

    Article  PubMed  Google Scholar 

  6. 6.

    Chen, Q. et al. Folate-PEG-NOTA-Al18F: A new folate based radiotracer for PET imaging of folate receptor-positive tumors. Mol. Pharm.14, 4353–4361 (2017).

    Article  CAS  Google Scholar 

  7. 7.

    Schniering, J. et al.18F-AzaFol for detection of folate receptor-β positive macrophages in experimental interstitial lung disease—A proof-of-concept study. Front. Immunol.10, 2724 (2019).

    Article  CAS  Google Scholar 

  8. 8.

    Müller, A. et al. Imaging atherosclerotic plaque inflammation via folate receptor targeting using a novel 18F-folate radiotracer. Mol. Imaging13, 1–11 (2014).

    Article  CAS  Google Scholar 

  9. 9.

    Gnesin, S. et al. Radiation dosimetry of 18F-AzaFol: A first in-human use of a folate receptor PET tracer. EJNMMI Res.10, 32 (2020).

    Article  CAS  Google Scholar 

  10. 10.

    Verweij, N. J. F. et al. First in man study of [18F]Fluoro-PEG-folate PET: A novel macrophage imaging technique to visualize rheumatoid arthritis. Sci. Rep.10, 1047 (2020).

    ADS  Article  CAS  Google Scholar 

  11. 11.

    Boss, S. D. et al. Reduced 18F-folate conjugates as a new class of PET tracers for folate receptor imaging. Bioconjug. Chem.29, 1119–1130 (2018).

    Article  CAS  Google Scholar 

  12. 12.

    Radford, L. L. et al. New 55Co-labeled albumin-binding folate derivatives as potential PET agents for folate receptor imaging. Pharmaceuticals12, 166 (2019).

    Article  Google Scholar 

  13. 13.

    Brand, C., Longo, V. A., Groaning, M., Weber, W. A. & Reiner, T. Development of a new folate-derived Ga-68-based PET imaging agent. Mol. Imaging Biol.19, 754–761 (2017).

    Article  CAS  Google Scholar 

  14. 14.

    Chen, Q. et al. Synthesis and preclinical evaluation of folate-NOTA-Al18F for PET imaging of folate-receptor-positive tumors. Mol. Pharm.13, 1520–1527 (2016).

    ADS  Article  CAS  Google Scholar 

  15. 15.

    Moisio, O. et al. Exploring alternative radiolabeling strategies for sialic acid-binding immunoglobulin-like lectin 9 peptide: [68Ga]Ga- and [18F]AlF-NOTA-Siglec-9. Molecules23, 305 (2018).

    Article  CAS  Google Scholar 

  16. 16.

    Feng, Y. et al. A folate receptor beta-specific human monoclonal antibody recognizes activated macrophage of rheumatoid patients and mediates antibody-dependent cell-mediated cytotoxicity. Arthritis Res. Ther.13, R59 (2011).

    Article  CAS  Google Scholar 

  17. 17.

    Stabin, M. G. & Siegel, J. A. RADAR dose estimate report: A compendium of radiopharmaceutical dose estimates based on OLINDA/EXM version 2.0. J. Nucl. Med.59, 154–160 (2018).

    Article  Google Scholar 

  18. 18.

    Parker, N. et al. Folate receptor expression in carcinomas and normal tissues determined by a quantitative radioligand binding assay. Anal. Biochem.338, 284–293 (2005).

    Article  CAS  Google Scholar 

  19. 19.

    Ayala-Lopez, W., Xia, W., Varghese, B. & Low, P. S. Imaging of atherosclerosis in apoliprotein e knockout mice: Targeting of a folate-conjugated radiopharmaceutical to activated macrophages. J. Nucl. Med.51, 768–774 (2010).

    Article  Google Scholar 

  20. 20.

    Winkel, L. C. et al. Folate receptor-targeted single-photon emission computed tomography/computed tomography to detect activated macrophages in atherosclerosis: Can it distinguish vulnerable from stable atherosclerotic plaques?. Mol. Imaging13, 1–5 (2014).

    Article  CAS  Google Scholar 

  21. 21.

    Pettinato, C. et al.68Ga-DOTANOC: biodistribution and dosimetry in patients affected by neuroendocrine tumors. Eur. J. Nucl. Med. Mol. Imaging.35, 72–79 (2008).

    Article  CAS  Google Scholar 

  22. 22.

    Kim, J. H. et al. Whole-body distribution and radiation dosimetry of 68Ga-NOTA-RGD, a positron emission tomography agent for angiogenesis imaging. Cancer Biother. Radiopharm.27, 65–71 (2012).

    Article  CAS  Google Scholar 

  23. 23.

    Roivainen, A. et al. Plasma pharmacokinetics, whole-body distribution, metabolism, and radiation dosimetry of 68Ga bombesin antagonist BAY 86–7548 in healthy men. J. Nucl. Med.54, 867–872 (2013).

    Article  CAS  Google Scholar 

  24. 24.

    Wurzer, A. et al. Molar activity of Ga-68 labeled PSMA inhibitor conjugates determines PET imaging results. Mol. Pharm.15, 4296–4302 (2018).

    Article  CAS  Google Scholar 

  25. 25.

    Spang, P., Herrmann, C. & Roesch, F. Bifunctional Gallium-68 chelators: Past, present, and future. Semin. Nucl. Med.46, 373–394 (2016).

    Article  Google Scholar 

  26. 26.

    De Sá, A. et al. Gallium labeled NOTA-based conjugates for peptide receptor-mediated medical imaging. Bioorganic Med. Chem. Lett.20, 7345–7348 (2010).

    Article  CAS  Google Scholar 

  27. 27.

    Notni, J., Pohle, K. & Wester, H.-J. Comparative gallium-68 labeling of TRAP-, NOTA-, and DOTA-peptides: Practical consequences for the future of gallium-68-PET. EJNMMI Res.2, 28 (2012).

    Article  CAS  Google Scholar 

  28. 28.

    Tsionou, M. I. et al. Comparison of macrocyclic and acyclic chelators for gallium-68 radiolabelling. RSC Adv.7, 49586–49599 (2017).

    Article  CAS  Google Scholar 

  29. 29.

    Skovgaard, D. et al. Safety, dosimetry and tumor detection ability of 68Ga-NOTA-AE105: first-in-humans study of a novel radioligand for uPAR PET imaging. J. Nucl. Med.58, 379–386 (2017).

    Article  CAS  Google Scholar 

Download references


The authors thank Aake Honkaniemi and Timo Kattelus for technical assistance. This work was financially supported by the Academy of Finland (Grant numbers 314553, 314554, and 314556), the Sigrid Jusélius Foundation, and the Jane and Aatos Erkko Foundation.

Author information




Conception and design (O.M., S.P., J.V., P.E., H.L., T.T., M.K., M.G.M., E.A.H., P.M., T.Ö., M.H., M.U.K., F.Z., M.S., J.K., P.S.L., A.S., X-G.L., A.R.), analysis and interpretation of data (O.M., S.P., J.V., P.E., T.T., M.K., M.G.M., E.A.H., P.M., T.Ö., M.H., M.U.K., F.Z., M.S., J.K., P.S.L., A.S., X-G.L., A.R.), drafting of the manuscript (O.M., S.P., J.V., P.E., T.T., F.Z, M.S., A.S., X-G.L., A.R.), revising it critically for important intellectual content (M.S., J.K., P.S.L., A.S., X-G.L., A.R.), and all authors have approved the final manuscript for submission.

Corresponding author

Correspondence to Anne Roivainen.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher's note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Moisio, O., Palani, S., Virta, J. et al. Radiosynthesis and preclinical evaluation of [68Ga]Ga-NOTA-folate for PET imaging of folate receptor β-positive macrophages. Sci Rep 10, 13593 (2020).

Download citation


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links