Radiolabelling and evaluation of a novel sulfoxide as a PET imaging agent for tumor hypoxia

Nuclear Medicine and Biology 41 (2014) 419–425

Contents lists available at ScienceDirect

Nuclear Medicine and Biology journal homepage: www.elsevier.com/locate/nucmedbio

Radiolabelling and evaluation of a novel sulfoxide as a PET imaging agent for tumor hypoxia Evelyn Laurens a, Shinn Dee Yeoh b, Angela Rigopoulos c, Diana Cao c, Glenn A. Cartwright c, Graeme J. O'Keefe b, Henri J. Tochon-Danguy b, d, Jonathan M. White a, Andrew M. Scott b, c, d, Uwe Ackermann b, c, d,⁎ a

School of Chemistry and Bio21 Institute, The University of Melbourne, Parkville VIC 3052, Australia Centre for PET, Austin Health, Level 1 HSB, 145 Studley Road, Heidelberg VIC 3084, Australia c Ludwig Institute for Cancer Research, Melbourne – Austin Branch, Heidelberg VIC 3084, Australia d School of Medicine, Dentistry and Health Sciences, The University of Melbourne, Melbourne VIC 3010, Australia b

a r t i c l e

i n f o

Article history: Received 13 June 2013 Received in revised form 26 February 2014 Accepted 1 March 2014 Keywords: Hypoxia Radiolabelling Click chemistry SK-RC-52 tumor model Radiotracer metabolism LCMS

a b s t r a c t [ 18F]FMISO is the most widely validated PET radiotracer for imaging hypoxic tissue. However, as a result of the pharmacokinetics of [18F]FMISO a 2 h wait between tracer administration and patient scanning is required for optimal image acquisition. In order to develop hypoxia imaging agents with faster kinetics, we have synthesised and evaluated several F-18 labelled anilino sulfoxides. In this manuscript we report on the synthesis, in vitro and in vivo evaluation of a novel fluoroethyltriazolyl propargyl anilino sulfoxide. The radiolabelling of the novel tracer was achieved via 2-[ 18F]fluoroethyl azide click chemistry. Radiochemical yields were 23 ± 4% based on 2-[ 18F]fluoroethyl azide and 7 ± 2% based on K[18F]F. The radiotracer did not undergo metabolism or defluorination in an in vitro assay using S9 liver fractions. Imaging studies using SKRC-52 tumors in BALB/c nude mice have indicated that the tracer may have a higher pO2 threshold than [ 18F] FMISO for uptake in hypoxic tumors. Although clearance from muscle was faster than [18F]FMISO, uptake in hypoxic tumors was slower. The average tumor to muscle ratio at 2 h post injection in large, hypoxic tumors with a volume greater than 686 mm3 was 1.7, which was similar to the observed ratio of 1.75 for [18F]FMISO. Although the new tracer showed improved pharmacokinetics when compared with the previously synthesised sulfoxides, further modifications to the chemical structure need to be made in order to offer significant in vivo imaging advantages over [ 18F]FMISO. © 2014 Elsevier Inc. All rights reserved.

1. Introduction Hypoxia plays an important role in many diseases such as diabetes, stroke and oncology and is a result of tissue oxygen consumption exceeding supply. In oncology, the abnormalities found in tumor vasculature cause insufficient blood supply to tumor cells and are responsible for the two major types of hypoxia found in tumors: chronic and acute hypoxia [1]. Chronic hypoxia is caused by a lack of blood vessels and is mainly found in tumor cells 50–200 μm away from the closest blood vessel [2]. Acute hypoxia shows a more diffused pattern within 50 μm of the nearest perfused vessel and is caused by sudden changes in blood flow. It is believed that both forms of hypoxia play an important role in the observed resistance of hypoxic tumors to chemotherapy and radiotherapy [3,4]. Numerous strategies have been developed to overcome therapy resistance of hypoxic tumors. Among those are the use of hypoxia ⁎ Corresponding author at: Centre for PET, Austin Health, Level 1 HSB, 145 Studley Road, Heidelberg VIC 3084, Australia. E-mail address: [email protected] (U. Ackermann). 0969-8051/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.nucmedbio.2014.03.001

directed cytotoxic drugs, radiosensitizers, intensity modulated radiotherapy, conformal radiotherapy or accelerated radiotherapy with carbogen and nicotinamide (ARCON) [5–7]. In order to accurately employ these therapeutic strategies, confirmation of the extent and location of hypoxia within the tumor is of critical importance. Measurement of oxygen partial pressure (pO2) in tumor tissue by means of a polarographic oxygen electrode is currently still the most accurate method however, this invasive technique is not suitable for routine clinical applications [8–10]. Positron emission tomography (PET) is a non-invasive nuclear medicine imaging technique that offers the potential to measure and quantify physiological processes in vivo. The most widely used PET imaging agents for tissue hypoxia in humans are [ 18F]FMISO and [ 18F] FAZA [11–15]. The retention of both radiotracers in hypoxic cells is based on the bioreduction of a nitroimidazole group and it has been demonstrated that [ 18F]FMISO shows uptake in tumors with a pO2 value of less than 10 mmHg [16–18]. However, slow accumulation in hypoxic tissue and slow clearance from normoxic tissue results in a low target to background ratio and a 2 h delay between tracer administration and the actual scanning of a patient needs to be observed [19]. This

420

E. Laurens et al. / Nuclear Medicine and Biology 41 (2014) 419–425

18

N

F

O

NO2

S

N

N

OH 18

F [18F]5

O 2N

18

[ F]FMISO

18

F

N O OH

OH

O S

N N

N

NO2

N

Cl

N [18F]FAZA

O2N

18

F

[18F]6

Fig. 1. Chemical structures of FMISO, FAZA and previously investigated sulfoxide imaging agents.

shortcoming has sparked the development of new tracers for hypoxia imaging. Most novel tracers have retained the 2-nitroimidazole core as the biologically relevant moiety responsible for the trapping in the hypoxic cell [20–23]. Bis(chloroethyl)-anilino sulfoxides are a class of hypoxia activated prodrugs that have the potential to be used as therapeutics as well as imaging agents [24]. These compounds rely on the bioreduction of a sulfoxide moiety followed by the formation of an aziridinium ion, which is generated due to the increased electron density at the anilino nitrogen after bioreduction. In our laboratory we have investigated radiolabelled anilino sulfoxides as a new class of hypoxia imaging agents [25,26]. Among those, the radioligands shown in Fig. 1, N-(2-[18F]fluoroethyl)-4-(4-nitrophenylsulfinyl)-N-(prop-2-ynyl)aniline ([18F]5) and N-(2-chloroethyl)-N-((1(2-[18F]fluoroethyl)-1H-1,2,3-triazol-4-yl)methyl)-4-(4-nitrophenylsulfinyl)benzeneamine ([18F]6) were of particular interest. Although [ 18F]5 showed uptake in an SK-RC-52 model of tumor hypoxia, clearance of this radioligand from normoxic tissue was slow. Despite bearing a chloroethyl group, [ 18F]6 was not retained in hypoxic tissue. In order to further investigate the properties of radiotracers bearing a propargyl group as potential hypoxia imaging agents, we synthesised 4-(4-nitrophenylsulfinyl)-N,N-(bisprop-2ynyl)aniline (4) and radiolabelled this precursor molecule with 2-[ 18F]fluoroethyl azide. The novel hypoxia radiotracer, N-(prop-2ynyl)-N-((1-(2-[ 18F]fluoroethyl)-1H-1,2,3-triazol-4-yl)methyl)-4(4-nitrophenylsulfinyl)benzeneamine ([ 18F]1), was investigated in vitro by measuring logD and performing stability studies in plasma as well as in S9 liver fractions. The pharmacokinetics of the radiotracer were evaluated in vivo by small animal PET imaging of BALB/c nude mice bearing transplanted SK-RC-52 tumors. Additionally, [ 18F] FMISO imaging of a selected number of those mice provided a direct comparison of the novel tracer with the [ 18F]FMISO standard. 2. Materials and methods 2.1. General No-carrier-added [ 18F]fluoride was produced by the 18O(p, n)18F nuclear reaction with a 10 MeV proton beam generated by the IBA Cyclone 10/5 cyclotron in a titanium target using [ 18O]H2O at Austin Health, Centre for PET. Typical irradiation parameters were 20 μA for

90 min, which produced 26.8–40.4 GBq (730–1100 mCi) of [ 18F] fluoride. Isolation of the [ 18F]fluoride ion from [ 18O]H2O was achieved by trapping on a QMA ion exchange column. Elution of the column with a solution containing 3.45 mg of anhydrous K2CO3 (0.025 mmol) and 20 mg of Kryptofix 2.2.2 (0.053 mmol) in 0.4 mL of acetonitrile plus 0.2 mL of water followed by repeated (3 times, 1.5 mL each) azeotropic evaporation with acetonitrile to dryness gave the anhydrous potassium [ 18F]fluoride complex used in the labelling experiments. Solvents were purchased from MERCK and used as received. Silica gel 60 (0.04–0.063 mm, MERCK) was used for flash chromatography and silica gel 60 F254 (0.2 mm, MERCK) was used for TLC analysis. Reagents were purchased from Sigma-Aldrich and used without further purification. 2-Azidoethyl 4-toluenesulfonate and 2-fluoroethyl 4-toluenesulfonate were synthesised according to literature procedures [27,28]. [ 18F]FMISO was synthesised from a commercially available precursor using the IBA Synthera module [15]. To determine HPLC retention time and confirm the identity of [19F]1, a Shimadzu 2010 LCMS system equipped with a 5 μL injection loop, a SPD-20A UV-Vis detector and two LC-20 AD solvent pumps for high pressure mixing of mobile phase were used. The stationary phase was a Phenomenex Gemini C-18, 5 μ RP column, 150 × 4.6 mm. Acetonitrile (A) and water (B) with 0.1% formic acid were used as the mobile phase at a flow rate of 0.5 mL/min and a gradient elution technique was used for analysis: 0–18 min: 5–90% A, 18–30 min: isocratic 90% A. For the quality control of [ 18F]1, the MS was replaced by the Bioscan FC-4000 dual BGO PET metabolite coincidence detector for the detection of radioactive compounds. Stationary phase and mobile phase composition was identical to those used with the LCMS system. Specific radioactivity was measured using a mass standard curve of known concentrations of [ 19F]1. For stability and metabolite studies, a Shimadzu HPLC system equipped with a 20 μL injection loop, a SPD-20A UV-Vis detector and two LC-20 AD solvent pumps for high pressure mixing of mobile phase were used. The stationary phase was a Phenomenex Gemini NX C-18, 5 μ RP column, 150 × 4.6 mm. Acetonitrile (A) and water (B) with 0.1% formic acid were used as the mobile phase at a flow rate of 0.5 mL/min and the following gradient elution technique was used for analysis: 0–18 min 5–90% A, 18–30 min isocratic 90% A. For the detection of radioactive compounds, the Bioscan FC-4000 dual BGO PET metabolite coincidence detector was used.

E. Laurens et al. / Nuclear Medicine and Biology 41 (2014) 419–425 1

H and 13C NMR spectra were recorded in CDCl3 on a Varian Unity 500 NMR spectrometer operating at 500 and 125 MHz respectively. High resolution mass spectrometry (HRMS) was carried out in the positive ion mode on a Finnigan LTQ-FT hybrid linear ion trap (Bremen, Germany) fitted with an electrospray ionisation (ESI) source and Fourier transform ion cyclotron resonance (FT-ICR). Infrared spectra (IR) were recorded on a Perkin Elmer Spectrum One, FT ATR-IR spectrometer. 2.2. Synthetic chemistry 2.2.1. 4-(4-Nitrophenylthio)-N,N-bis(prop-2-ynyl)aniline (3) To a stirred solution of 4-(4-nitrophenylthio)aniline (0.50 g, 2.03 mmol) in acetonitrile (50 mL), were added propargyl bromide (80%, 1.28 g, 0.92 mL, 8.12 mmol, 4 eq.) and 0.83 g of anhydrous potassium carbonate (6 mmol). The mixture was refluxed under N2 for 19 h. The reaction mixture was then cooled to room temperature, diluted with water (30 mL) and extracted with diethyl ether (3 × 30 mL). The organic phases were combined, washed with NaHCO3 (3 × 50 mL), H2O (3 × 50 mL), dried (MgSO4) and concentrated under reduced pressure to give a bright yellow residue. The residue was purified by flash chromatography on silica gel (gradient 2–80% ether in hexane) followed by recrystallization with dichloromethane/n-pentane to give 3 as bright yellow solid (188 mg, 33% yield) m.p 93.5–93.9 °C; TLC Rf 0.67 in ether/hexane (1:1). 1

H NMR : 2:30ð2H; s; C≡CHÞ; 4:19ð4H; d; J ¼ 2Hz; N  CH2 Þ; 6:99ð2H; d; J ¼ 8:5Hz; CH  ArÞ; 7:11ð2H; d; J ¼ 9:0Hz; CH  ArÞ; 7:47ð2H; d; J 13 ¼ 9:0Hz; CH  ArÞ; 8:04ð2H; d; J ¼ 9:0Hz; CH  ArÞ; CNMR : 40:25ðCH2  NÞ; 72:98ðCH≡CÞ; 78:59ðCH2  C≡CHÞ; 115:74ðCH  ArÞ; 117:71ðC  S; ArÞ; 123:91ðCH  ArÞ; 125:50ðCH  ArÞ; 136:83ðCH  ArÞ; 148:78ðC  N; ArÞ; 150:49ðC  N; ArÞ; IR υmax : 1328:9; 1498:38ðNO2 Þ; 2113:7ðC≡CÞ; 3096:3ðC  HÞ; 3278:6ð≡C −1  HÞcm ; HRMS þ : m=zðESIþÞ323:08485ðC18 H14 N2 O2 SH½M þ H Calcd 323:08487Þ

2.2.2. 4-(4-Nitrophenylsulfinyl)-N,N-bis(prop-2-ynyl)aniline (4) A solution of NaIO4 (125 mg, 0.586 mmol, 1 eq.) in H2O (1 mL) was added to a solution of 3 (160 mg, 0.586 mmol, 1 eq.) in methanol (30 mL) and was refluxed under N2 for 18 h. The reaction was cooled to room temperature, diluted with water (30 mL) and extracted with dichloromethane (3 × 30 mL). The organic phases were combined, washed with water (50 mL), dried (MgSO4) and concentrated under reduced pressure to give a yellow residue. The residue was purified by flash chromatography on silica gel (gradient 2–80% diethyl ether in hexane) followed by recrystallization with dichloromethane/npentane to give 4 as bright yellow solid (101 mg, 51% yield) m.p 120.3–121.2 °C; TLC Rf 0.28 in ether/hexane (7:3). 1

H NMR : 2:26ð2H; m; C≡CHÞ; 4:14ð4H; d; J ¼ 2:5Hz; N  CH2 Þ; 6:95ð2H; d; J ¼ 9:0Hz; CH  ArÞ; 7:54ð2H; d; J ¼ 8:0Hz; CH  ArÞ; 7:78ð2H; d; J 13 ¼ 8:5Hz; CH  ArÞ; 8:30ð2H; d; J ¼ 8:5Hz; CH  ArÞ; CNMR : 40:23ðCH2  NÞ; 70:12ðCH≡CÞ; 78:16ðCH2  C≡CHÞ; 114:61ðCH  ArÞ; 124:21ðCH  ArÞ; 125:36ðCH  ArÞ; 127:64ðCH  ArÞ; 132:64ðC  S; ArÞ; 149:05ðC  S; ArÞ; 150:30ðC  N; ArÞ; 153:46ðC  N; ArÞ; IR υmax : 1049:8ðS ¼ OÞ; 1337:8; 1506:5ðNO2 Þ; 2114:6ðC≡CÞ; 3100:6ðC −1  HÞ; 3279:9ð≡C  HÞcm ; HRMS þ : m=zðESIþÞ339:07974ðC18 H14 N2 O3 SH½M þ H Calcd 339:07979Þ

2.2.3. 2-[ 19F]Fluoroethyl azide Sodium azide (65 mg, 1 mmol) was added to a solution of 2-fluoroethyl-4-toluenesulfonate (218 mg, 1 mmol) in 10 mL of

421

anhydrous DMF. The mixture was stirred at room temperature for 24 h to give 2-[ 19F]fluoroethyl azide, which was used for the subsequent click reaction without further purification. 2.2.4. Preparation of the copper catalyst for click chemistry H2O (250 μL) was added to CuI (2.5 mg, 0.013 mmol) and sodium ascorbate (22.9 mg, 0.13 mmol). Diisopropylethylamine (25 μL, 0.14 mmol), DMF (250 μL) and acetonitrile (250 μL) were then added and this mixture was used to catalyse both the F-18 und F-19 click reactions. 2.2.5. N-(Prop-2-ynyl)-N-((1-(2-[ 19F]fluoroethyl)-1H-1,2,3-triazol-4-yl) methyl)-4-(4-nitrophenylsulfinyl)benzeneamine (1) To a stirred solution of 4 (40 mg, 0.12 mmol) in 1 mL of acetonitrile, was added the prepared copper catalyst mixture. After the addition of the 2-[ 19F]fluoroethyl azide solution (0.14 mL, 0.14 mmol), the reaction was stirred at room temperature for 48 h. Reversed phase C-18 material (2 g) was then added to the reaction mixture and the solvent was removed using reduced pressure. Using the Reveleris™ solid loader, [ 19F]1 was then purified by preparative reversed phase C-18 chromatography with the Reveleris™ system. A 12 g Reveleris™ reversed phase C-18 cartridge and a water(A)/ acetonitrile(B) gradient system (33 min: 0–50%B, 33–39 min: 50– 100%B, 39–40 min: 100%B) at a flow rate of 20 mL/min was used to elute the target compound. The retention time of [ 19F]1 under those conditions was 27.1 min. After evaporation of the solvent, [ 19F]1 was obtained as yellow oil (10.1 mg, 22% yield). 1

H NMR : 2:26ð1H; m; C≡CHÞ; 4:16ð2H; d; J ¼ 2:5Hz; CH2 Þ; 4:62ð2H; dt; J ¼ 27Hz; 5Hz; CH2 Þ; 4:73ð2H; s; CH2 Þ; 4:76ð2H; dt; J ¼ 48Hz; 5Hz; CH2 Þ; 6:93ð2H; d; J ¼ 9:0Hz; CH  ArÞ; 7:49ð2H; d; J ¼ 9Hz; CH  ArÞ; 7:54ð1H; s; CH  ArÞ; 7:77ð2H; d; J 13 ¼ 9Hz; CH  ArÞ; 8:29ð2H; d; J ¼ 9Hz; CH  ArÞ; CNMR : 40:51ðN  CH2 Þ; 46:94ðN  CH2 Þ; 50:61ðd; J ¼ 20:3Hz; CH2  NÞ; 73:02ðC≡CHÞ; 78:56ðC≡CHÞ; 81:45ðd; J ¼ 172:9Hz; CH2  FÞ; 113:85ðCH  ArÞ; 124:17ðCH  ArÞ; 125:35ðCH  ArÞ; 127:88ðCH  ArÞ; 129:91ðCH  ArÞ; 131:79ðC  S; ArÞ; 144:59ðC  ArÞ; 148:99ðCS  ArÞ; 150:37ðC  N; ArÞ; 153:45ðC  N; ArÞ; HRMS : m=zðESIþÞ428:11863ðC20 H18 FN5 O3 SH½M þ þ H Calcd 428:11871Þ

2.3. Radiochemistry 2.3.1. 2-[ 18F]Fluoroethyl azide To the dried [ 18F]KF/Kryptofix complex was added 10 μL (12 mg, 0.05 mmol) of 2-azidoethyl 4-toluenesulfonate in 400 μL of acetonitrile. The reaction mixture was then heated in a sealed reaction vial to 110 ºC. After 15 min, 600 μL of acetonitrile were added and the temperature was raised to 140 ºC. 2-[ 18F]Fluoroethyl azide was then distilled into a separate reaction vial which contained 4-(4-nitrophenylsulfinyl)-N,N-bis(prop-2-ynyl)aniline (4) (2 mg, 6.7 μmol) in 150 μL of acetonitrile. The distillation was performed using a constant nitrogen flow of 20 mL/min, which was bubbled through the solution. 2.3.2. N-(Prop-2-ynyl)-N-((1-(2-[ 18F]fluoroethyl)-1H-1,2,3-triazol-4-yl) methyl)-4-(4-nitrophenylsulfinyl)benzeneamine ([18F]1) After 2-[ 18F]fluoroethyl azide radioactivity had reached its maximum in the second reactor vial, the freshly prepared solution of the copper catalyst was added to the 2-[ 18F]fluoroethyl azide/ acetylene solution. The reaction mixture was then heated to 70 ºC for 15 min, followed by the addition of 3 mL of H2O and injection into a semi-preparative HPLC system for purification. Semi-preparative

422

E. Laurens et al. / Nuclear Medicine and Biology 41 (2014) 419–425

HPLC was performed using a Shimadzu LC-10AS isocratic pump equipped with a 5 mL injection loop and a reversed phase column (Alltech Apollo™ C-18, 5 μ, 10 × 250 mm). A 40/60 0.1 M ammonium formate/acetonitrile mixture at a flow rate of 4 mL/min was used as a mobile phase. Detection of chemical compounds was achieved with a Shimadzu SPD-6AV UV detector at 254 nm and a Geiger-Müller tube (type 716; LND, Inc., NY), as radiodetector. The radioactive peak at 14.5 min was collected and reformulated using the solid phase extraction (SPE) method [29]. 2.4. In vitro and in vivo evaluation 2.4.1. Stability in saline, human plasma and S9 liver fractions The in vitro stability assays in saline, human plasma and S9 liver fractions were performed as previously described by us [26,30]. 2.4.2. LogD measurement LogD was measured by mixing 3.7 MBq of the radiotracer with 1 g each of 1-octanol and phosphate buffer (0.1 M, pH 7.4) in a test tube. The test tube was vortexed for 3 min at room temperature, followed by centrifugation for 5 min at 11,000 rcf. Two weighed samples (0.5 g each) from the 1-octanol and buffer layers were then measured for radioactivity. The distribution coefficient was calculated from the ratio of cpm/g of 1-octanol to that of buffer. 2.5. Animal experiments All animal experiments were approved by the Austin Health animal ethics committee. 2.5.1. Tumor transplants SK-RC-52 tumors were first grown by injecting suspensions of 6 × 10 6 SK-RC-52 cells subcutaneously into the flank of BALB/c nude mice. Tumors were allowed to grow to a size of about 300 mm 3, then cut into 40 mm 3 pieces and transplanted into the shoulder of BALB/c nude mice. Imaging started when the tumors were 180 mm 3 in size. The maximum tumor size imaged was 994 mm 3. 2.5.2. PET imaging Imaging studies were performed using a Mosaic small animal PET scanner. Animals were injected with 9.25 MBq of radiotracer in 100 μL of the final formulation and anaesthetized using isoflurane delivered by the Minerva Biovet animal imaging system before scanning. The Minerva Biovet system also provides a temperature stabilised environment for the animals during the induction and imaging stages. Dynamic imaging was performed by acquiring 12 × 10 min frames from the start of tracer injection. For all mice, a 10 min static frame was acquired 2 h post injection. Images were reconstructed using the RAMLA3D algorithm [31]. Following PET image acquisition, the animals were relocated to a Philips Gemini PET/ CT and CT scanned at 90 kVp with a 150 mA tube current at 0.5 s per rotation. The acquired CT had a pixel size of 0.098 mm and a slice thickness of 0.6 mm.

optical oxygen-sensing probe (Oxylite 2000; Oxford Optronix, Oxford, UK) or oxylite probe. The probes (230 μm o.d.) were precalibrated by the manufacturer (± 0.7 mmHg or b ±10% of actual pO2, whichever was greater). To further ensure correct pO2 readings in the experiments, the probe was checked in normal saline, and again in animals just sacrificed to ensure a 0 mmHg recording. 3. Results and discussion 3.1. Synthetic chemistry Fig. 2 shows the synthesis of the precursor and the labelling of the radioligand as well as the cold standard synthesis. The precursor 4 could be prepared in an overall chemical yield of 18% from commercially available 4-(4-nitrophenylthio)aniline in 2 steps. The F-19 standard was prepared from 4 using in situ generated 2-[ 19F] fluoroethyl azide and a Cu(I) catalyst to facilitate the click chemistry reaction. Stirring of the reaction mixture at room temperature for 48 h gave [ 19F]1 in 22% yield. 3.2. Radiochemistry 3.2.1. Synthesis of [ 18F]1 4-(4-Nitrophenylsulfinyl)-N,N-bis(prop-2-ynyl)aniline (4) was reacted with 2-[ 18F]fluoroethyl azide and a Cu(I) catalyst at 70 ºC for 15 min. The radiotracer [ 18F]1 was purified via semi-preparative HPLC and reformulated in 10% ethanol/saline. Radiochemical yields were 7 ± 2% based on K[ 18F]F. Radiochemical purity was N96% and specific activity at the end of synthesis (EOS) ranged from 75.2 to 134.8 GBq/μmol. The synthesis time including HPLC purification was 160 min. Fig. 3 shows HPLC and LCMS chromatograms of the quality control. HPLC analysis was performed on a Shimadzu HPLC system using an analytical Phenomenex Gemini™ reverse phase C-18 column and a H2O/CH3CN gradient solvent system as mobile phase. The retention time of [ 18F]1 was 18.5 min. Identity of the tracer was confirmed by comparison with the cold standard. 3.3. In vitro studies 3.3.1. Lipophilicity The octanol/water coefficient was measured and the logD was calculated to be 1.45 for [ 18F]1, thus indicating that the lipophilicity of this compound is suitable for crossing cell membranes. 3.3.2. Stability and metabolite studies The radiotracer was found to be stable in human plasma and saline for a period of 2 h. A cytochrome P450 assay using mouse S9 liver fractions also showed no metabolism (Fig. 4) or defluorination of [ 18F]1. 3.4. In vivo evaluation of [ 18F]1

2.5.3. Image analysis The resultant PET and CT images where imported into the PMOD analysis system for spatial alignment. Regions of interest for tumor and muscle tissue were generated and tumor to muscle ratios were calculated by dividing the average counts of the 5 hottest pixels in tumor by the mean radioactivity in the reference area [32]. 2.5.4. Oxygen tension measurement After imaging, animals selected for pO2 measurement were humanely euthanized and the oxygen partial pressure in the tumor was measured using a polarographic oxygen electrode. Tumor pO2 was measured with a 2-channel time-resolved luminescence-based

In vivo evaluation of [ 18F]1 was carried out using transplanted SKRC-52 tumors as a model for tumor hypoxia [33,34]. Tumors were grown in the shoulder of BALB/c nude mice and tumor sizes were measured using a calliper. Animals were imaged with [ 18F]FMISO and the novel tracer at different stages of tumor growth, ranging from 180 mm 3 to 994 mm 3. For all animals, a 10 min static image at 2 h post tracer administration was acquired. This image was used to measure the tumor to muscle (t/m) ratio at this timepoint. For pharmacokinetic measurements, dynamic imaging of animals over a period of 2 h (12 × 10 min frames) was performed and time activity curves generated for tumor and muscle tissue.

E. Laurens et al. / Nuclear Medicine and Biology 41 (2014) 419–425

S

NH2

423

S

N

i

O2N

O 2N

3 ii

2

O S

N

O2N

4 iv

iii

N N O S

O

N

S

N

N

N N

N

19 18

O 2N

[18F]1

O 2N

F

F [19F]1

Fig. 2. Precursor synthesis and radiolabelling reactions. Reagents and conditions: i: 4 eq. propargyl bromide, K2CO3, CH3CN, reflux, 19h; ii: NaIO4, MeOH, reflux, 18 h; iii: 2-[18F] fluoroethylazide, CuI, sodium ascorbate, 70ºC, 15 min; iv: 2-[18F]fluoroethylazide CuI, sodium ascorbate, RT, 48 h.

For [ 18F]1, 2 small, 2 medium and 4 large size tumors were imaged statically, and 3 of the mice bearing large tumors were also imaged dynamically. No tumor uptake was observed when the size of the tumor was between 180 mm 3 and 188 mm 3 (average size 184 mm 3, n = 2). The pO2 level in the 188 mm 3 tumor was measured using a polarographic oxygen probe and was found to be greater than 20 mmHg. Tumors in this group were therefore considered not to be hypoxic. When the tumor size was between 360 and 390 mm3 (average size 375 mm3, n = 2) we observed uptake of the tracer in this small cohort of animals. The mean pO2 value in the 390 mm3 tumor was 12 mmHg and tumors were therefore considered to be mildly hypoxic.

3.5

mV(x100) AD2

In tumors ranging from 686 to 838 mm 3 (average tumor size 764 mm 3, n = 4) the average t/m ratio was 1.7 ± 0.3. One animal bearing a 773 mm 3 tumor was sacrificed after imaging and the measured pO2 value of 4 mmHg confirmed the presence of hypoxia. Dynamic imaging of [ 18F]1 of 3 mice over a period of 2 h revealed that the tracer accumulated within the tumor over the 2 h time period without reaching equilibrium (Fig. 5). 3.4.1. Imaging studies and pharmacokinetic comparison between [ 18F]1 and [ 18F]FMISO Fig. 5 illustrates [ 18F]1 imaging studies and pharmacokinetic comparison to [ 18F]FMISO. For [ 18F]FMISO, 1 small (236 mm 3),

2.00

3.0

1.75

2.5

1.50

mV(x10) Detector A:254nm

1.25

2.0

1.00

1.5

0.75

1.0

0.50

0.5

0.25 0.00

0.0 0.0

2.5

5.0

7.5

10.0

12.5

15.0

17.5

20.0

22.5

25.0

27.5

min

0.0

2.5

5.0

7.5

[18F]1 radioactivity channel 1.00

10.0

12.5

15.0

17.5

20.0

22.5

25.0

27.5

min

[18F] 1 UV channel

mV(x1,000) SPD-UV:254nm

(x10,000,000) 427.95 (1.21)

1.50 0.75

1.25 1.00

0.50

0.75 0.50

0.25

0.25 0.00

0.00 0.0

2.5

5.0

7.5

10.0

12.5

15.0

17.5

20.0

[ 19F]1 UV channel

22.5

25.0

27.5

min

0.00

2.5

5.0

7.5

10.0

12.5

15.0

17.5

20.0

22.5

25.0

27.5

30.0

[19F] 1 LCMS molecular ion [M+H]+ = m/z 428

Fig. 3. Quality control of [18F]1 (radioactivity: top left, UV: top right) and comparison to cold standard [19F]1 (UV: bottom left, m/z = 428: bottom right).

424

E. Laurens et al. / Nuclear Medicine and Biology 41 (2014) 419–425 mV(x100) RAD

4.0

1.25

mV(x10) RAD

3.5 3.0

1.00

2.5 0.75

2.0 1.5

0.50

1.0

0.25

0.5 0.0

0.00 0.0

5.0

10.0

15.0

20.0

25.0

30.0

min

0.0

5.0

10.0

control

15.0

20.0

25.0

30.0

min

120 min incubation

Fig. 4. Radio HPLC chromatograms showing the stability of [18]F1 towards S9 liver fractions over 2 h; control at t = 0 min (left) vs 120 min S9 liver fractions incubation (right).

4500

Average counts per voxel

Average counts per voxel

4500 4000 3500 3000

tumor muscle

2500 2000 1500 1000

0

20

40

60

80

100

120

4000 3500 3000

tumor muscle

2500 2000 1500 1000

0

20

time (min)

40

60

80

100

120

time (min)

Fig. 5. Averaged time activity curves of [18F]1 (left) and [18F]FMISO (right) in hypoxic, SK-RC-52 tumor bearing mice.

1 medium (328 mm 3) and 2 large (942, 994 mm 3) tumors were imaged statically. Dynamic image acquisition was also performed on the animals bearing the 2 large tumors. As expected, no tumor uptake was observed in small and medium size tumors with [ 18F]FMISO. The average t/m ratio for the large tumors at 2 h post injection was 1.75 ± 0.05, which was slightly higher than what was observed for [ 18F]1. Fig. 6 shows comparison images of 10 min static scans between [ 18F]FMISO and [ 18F]1 acquired at 2 h post injection.

tumor

Dynamic imaging showed that the average clearance of [ 18F]1 from muscle was 25 ± 5.7%, whereas the average clearance from muscle for [ 18F]FMISO was only 14.4 ± 6% over the 2 h imaging period. However, the accumulation in hypoxic tumors was slower for [ 18F]1 (150.7 ± 7%) than the accumulation for [ 18F]FMISO (236 ± 63.5%) over 2 h. From the time activity curves obtained for [ 18F]1 we conclude that the presence of a polar fluoroethyl triazolyl moiety is beneficial for fast clearance from normoxic tissue. In hypoxic tissue, the sulfoxide

tumor

Fig. 6. Representative coronal images acquired at 120 min post injection, 10 min static scan, of [18F]1 (left) and [18F]FMISO (right) in 760, 773 and 942 and 994 mm3 tumors, respectively.

E. Laurens et al. / Nuclear Medicine and Biology 41 (2014) 419–425

group is reduced to the thioether and the propargyl group can undergo rearrangement to a reactive allene species, which is believed to be responsible for the trapping of the radiotracer [26]. Our results also indicated that [ 18F]1 has a different hypoxia selectivity from [ 18F]FMISO since we have observed uptake in a small cohort of animals bearing tumors with a pO2 value greater than 10 mmHg. Overall, the novel radioligand shows uptake in hypoxic tissue and fast clearance from normoxic tissue. However, modifications to the chemical structure need to be made to improve the rate of uptake in hypoxic tissue.

[12]

[13] [14] [15] [16]

[17]

4. Conclusion [18]

We have synthesised a novel F-18 labelled 4-nitrophenyl sulfoxide, [ 18F]1, and evaluated this compound in vitro and in vivo. The tracer was stable towards metabolism by S9 liver fractions and has a lipophilicity that is suitable for crossing cell membranes. Our in vivo investigations have shown that [ 18F]1 has slightly higher hypoxia selectivity than [ 18F]FMISO. Although the rate of clearance of [ 18F]1 from normoxic tissue was faster than that of [ 18F]FMISO, the rate of uptake of [ 18F]1 in hypoxic tumors was slower. As a result, the t/m ratio of [ 18F]1 in hypoxic SK-RC-52 tumors is slightly lower than that of [ 18F]FMISO in tumors of similar size. Due to its fast clearance from normoxic tissue, [ 18F]1 is an exciting lead structure for the future development of hypoxia imaging agents.

[19]

[20]

[21]

[22]

Acknowledgments

[23]

This work was supported in part by grants from the National Health and Medical Research Council (NHMRC) 469002 and 487922.

[24]

[25]

References [1] Bussink J, Kaanders JHAM, van der Kogel AJ. Tumor hypoxia at the microregional level: clinical relevance and predictive value of exogenous and endogenous hypoxic cell markers. Radiother Oncol 2003;67:3–15. [2] Olive PL, Durand RE, Raleigh JA, Luo C, Aquino-Parsons C. Comparison between the comet assay and pimonidazole binding for measuring tumor hypoxia. Br J Cancer 2000;83:1525–31. [3] Lara PC, Lloret M, Clavo B, Apolinario RM, Henriquez-Hernandez LA, Bordón E, et al. Severe hypoxia induces chemo-resistance in clinical cervical tumors through MVP overexpression. Radiat Oncol 2009;4:29–33. [4] Sasabe E, Zhou X, Li D, Oku N, Yamamoto T, Osaki T. The involvement of hypoxia inducible factor-1 alpha in the susceptibility to gamma rays and chemotherapeutic drugs of oral squamous cell carcinoma cells. Int J Cancer 2007;120:268–77. [5] Chapman JD, Franko AJ, Sharplin J. A marker for hypoxic cells in tumours with potential clinical applicability. Br J Cancer 1981;43:546–50. [6] Hoskin PJ, Saunders MI. ARCON. Clin Oncol (R Coll Radiol) 1994;6:281–2. [7] Rajendran JG, Hendrickson KRG, Spence AM, Muzi M, Krohn KA, Mankoff DA. Hypoxia imaging-directed radiation treatment planning. Eur J Nucl Med Mol Imaging 2006;33:S44–53. [8] Stone HB, Brown JM, Phlilips TL, Sutherland RM. Oxygen in human tumors: correlations between methods of measurements and response to therapy. Radiat Res 1993;136:422–34. [9] Vaupel P, Schlenger K, Knoop C, Höckel M. Oxygenation of human tumors: evaluation of tissue oxygenation distribution in breast cancers by computerized O2 tension measurements. Cancer Res 1991;51:3316–22. [10] Höckel M, Knoop C, Schlenger K, Vorndran B, Baussmann E, Mitze M, et al. Intratumoral pO2 predicts survival in advanced cancer of the uterine cervix. Radiother Oncol 1993;26:45–50. [11] Markus R, Donnan GA, Kazui S, Read S, Hirano T, Scott AM, et al. Statistical parametric mapping of hypoxic tissue identified by [(18)F]fluoromisonidazole

[26]

[27] [28] [29]

[30]

[31]

[32]

[33]

[34]

425

and positron emission tomography following acute ischemic stroke. Neuroimage 2002;16:425–33. Brady F, Luthra KS, Brown GD, Osman S, Aboagye E, Saleem A, et al. Radiolabelled tracers and anticancer drugs for assessment of therapeutic efficacy using PET. Curr Pharm Des 2001;7(18):1863–92. Nunn A, Linder K, Strauss HW. Nitroimidazoles and imaging hypoxia. Eur J Nucl Med 1995;22:265–80. Foo SS, Abbott DF, Lawrentschuk N, Scott AM. Functional imaging of intratumoural hypoxia. J Mol Imaging Biol 2004;6(5):291–305. Lee ST, Scott AM. Hypoxia positron emission tomography imaging with 18Ffluoromisonidazole. Semin Nucl Med 2007;37(6):451–61. Gagel B, Reinartz P, DiMartino E, Zimny M, Pinkawa M, Maneschi P, et al. pO2 polarography versus positron emission tomography ([18F] fluoromisonidazole, [18F]-2-fluoro-2′-deoxyglucose). Strahlenther Onkol 2004;10:616–22. Chang JH, Wada M, Anderson NJ, Joon DL, Lee ST, et al. Hypoxia-targeted radiotherapy dose painting for head and neck cancer using 18F-FMISO PET: a biological modeling study. Acta Oncol 2013;52:1723–9. Hendrickson K, Phillips M, Smith W, Peterson L, Krohn K, et al. Hypoxia imaging with [F-18] FMISO-PET in head and neck cancer: potential for guiding intensity modulated radiation therapy in overcoming hypoxia-induced treatment resistance. Radiother Oncol 2011;101:369–75. Read SJ, Hirano T, Abbott DF, Markus R, Sachinidis JI, Tochon-Danguy HJ, et al. The fate of hypoxic tissue on 18F-fluoromisonidazole positron emission tomography after ischemic stroke. Ann Neurol 2000;48:228–35. Dubois L, Landuyt W, Cloetens L, Bol A, Bormans G, Haustermans K, et al. [18F]EF3 is not superior to [18F]FMISO for PET-based hypoxia evaluation as measured in a rat rhabdomyosarcoma tumour model. Eur J Nucl Med Mol Imaging 2009;36:209–18. van Loon J, Janssen MHM, Öllers M, Aerts HJWL, Dubois L, Hochstenbag M, et al. PET imaging of hypoxia using [18F]HX4: a phase I trial. Eur J Nucl Med Mol Imaging 2010;37:1663–8. Koch CJ, Scheuermann JS, Divgi C, Judy KD, Kachur AV, Freifelder R, et al. Biodistribution and dosimetry of 18F-EF5 in cancer patients with preliminary comparison of 18F-EF5 uptake versus EF5 binding in human glioblastoma. Eur J Nucl Med Mol Imaging 2010;37:2048–59. Zha Z, Zhu L, Liu Y, Du F, Gan H, Qiao J, et al. Synthesis and evaluation of two novel 2-nitroimidazole derivatives as potential PET radioligands for tumor imaging. Nucl Med Biol 2011;38:501–8. Sun ZY, Botros E, Su AD, Kim Y, Wang E, Baturay NZ, et al. Sulfoxide-containing aromatic nitrogen mustards as hypoxia-directed bioreductive cytotoxins. J Med Chem 2000;43:4160–8. Falzon CL, Ackermann U, Spratt N, Tochon-Danguy HJ, White J, Howells D, et al. F18 labelled N,N-bis-haloethylamino-phenylsulfoxides — a new class of compounds for the imaging of hypoxic tissue. J Label Compd Radiopharm 2006;49:1089–103. Laurens E, Yeoh SD, Rigopoulos A, Cao D, Cartwright GA, O'Keefe GJ, et al. Radiolabelling and evaluation of novel haloethylsulfoxides as PET imaging agents for tumor hypoxia. Nucl Med Biol 2012;39:871–82. Demko ZP, Sharpless KB. An intramolecular [2 + 3] cycloaddition route to fused 5-heterosubstituted tetrazoles. Org Lett 2001;25:4091–4. Studenov AR, Berridge MS. Synthesis and properties of 18F-labeled potential myocardial blood flow tracers. Nucl Med Biol 2001;28:683–93. Lemaire C, Plenevaux A, Aerts J, Del Fiore A, Brihaye C, Le Bars D, et al. Solid phase extraction — an alternative to the use of rotary evaporators for solvent removal in the rapid formulation of PET radiopharmaceuticals. J Label Compd Radiopharm 1999;42:63–75. Ackermann U, Sigmund D, Yeoh SD, Rigopoulos A, O'Keefe G, Cartwright G, et al. Synthesis of 2-[(4-[18F]fluorobenzoyloxy)methyl]-1,4-naphthalenedione from 2hydroxymethyl 1,4-naphthoquinone and 4-[18F]fluorobenzoic acid using DCC. J Label Compd Radiopharm 2011;54:788–94. Browne J, de Pierro AB. A row-action alternative to the EM algorithm for maximizing likelihood in emission tomography. IEEE Trans Med Imaging 1996;15 (5):687–99. Pantaleo MA, Mishani E, Nanni C, Landuzzi L, Boschi S, Nicoletti G, et al. Evaluation of modified PEG-anilinoquinazoline derivatives as potential agents for EGFR imaging in cancer by small animal PET. Mol Imaging Biol 2010;12:616–25. Lawrentschuk N, Poon AM, Foo SS, Putra LG, Murone C, Davis ID, et al. Assessing regional hypoxia in human renal tumours using 18F-fluoromisonidazole (18FFMISO) positron emission tomography. BJU Int 2005;96:540–6. Lawrentschuk N, Lee FT, Jones G, Rigopoulos A, Mountain A, O'Keefe G, et al. Investigation of hypoxia and carbonic anhydrase IX expression in a renal cell carcinoma xenograft model with oxygen tension measurements and 124I-cG250 PET/CT. Urol Oncol 2011;29(4):411–20.