Available online at www.sciencedirect.com
Nuclear Medicine and Biology 38 (2011) 235 – 245 www.elsevier.com/locate/nucmedbio
Synthesis of hypoxia imaging agent 1-(5-deoxy-5-fluoro-α-D-arabinofuranosyl)-2-nitroimidazole using microfluidic technology Vincent R. Bouvet, Melinda Wuest, Leonard I. Wiebe, Frank Wuest⁎ Department of Oncology, University of Alberta, Edmonton, AB, Canada T6G 1Z2 Received 5 July 2010; received in revised form 2 September 2010; accepted 6 September 2010
Abstract Introduction: Microfluidic technology allows fast reactions in a simple experimental setup, while using very low volumes and amounts of starting material. Consequently, microfluidic technology is an ideal tool for radiolabeling reactions involving short-lived positron emitters. Optimization of the complex array of different reaction conditions requires knowledge of the different reaction parameters linked to the microfluidic system as well as their influence on the radiochemical yields. 1-(5-Deoxy-5-fluoro-α-D-arabinofuranosyl)-2-nitroimidazole ([18F]FAZA) is a frequently used radiotracer for PET imaging of tumor hypoxia. The present study describes the radiosynthesis of [18F] FAZA by means of microfluidic technology and subsequent small animal PET imaging in EMT-6 tumor-bearing mice. Methods: Radiosyntheses were performed using the NanoTek Microfluidic Synthesis System (Advion BioSciences, Inc.). Optimal reaction conditions were studied through screening different reaction parameters like temperature, flow rate, residency time, concentration of the labeling precursor (1-(2,3-di-O-acetyl-5-O-tosyl-α-D-arabinofuranosyl)-2-nitroimidazole) and the applied volume ratio between the labeling precursor and [18F]fluoride. Results: Optimized reaction conditions at low radioactivity levels (1 to 50 MBq) afforded 63% (decay-corrected) of HPLC-purified [18F] FAZA within 25 min. Higher radioactivity levels (0.4 to 2.1 GBq) gave HPLC-purified [18F]FAZA in radiochemical yields of 40% (decay-corrected) within 60 min at a specific activity in the range of 70 to 150 GBq/μmol. Small animal PET studies in EMT-6 tumorbearing mice showed radioactivity accumulation in the tumor (SUV20min 0.74 ± 0.08) resulting in an increasing tumor-to-muscle ratio over time. Conclusions: Microfluidic technology is an ideal method for the rapid and efficient radiosynthesis of [18F]FAZA for preclinical radiopharmacological studies. Careful analysis of various reaction parameters is an important requirement for the understanding of the influence of different reaction parameters on the radiochemical yield using microfluidic technology. Exploration of microfluidic technology for the radiosynthesis of other PET radiotracers in clinically relevant radioactivity levels is currently in progress. © 2011 Elsevier Inc. All rights reserved. Keywords: [18F]FAZA; Microfluidic technology; EMT-6 tumor-bearing mice
1. Introduction In recent years, various micro-reactor and microfluidic devices have emerged as novel and highly innovative technology for the synthesis of radiolabeled compounds. The benefit of micro-reactor and microfluidic technology in chemical synthesis, including radiochemistry, was convinc-
⁎ Corresponding author. Tel.: +1 780 989 8150; fax: +1 780 432 8483. E-mail address:
[email protected] (F. Wuest). 0969-8051/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.nucmedbio.2010.09.002
ingly demonstrated [1–7]. Application of micro-reactor and microfluidic technology offers major advantages like higher yields, shorter reaction times and purer product formation compared to conventional batch reactions. Miniaturization of radiosynthesis devices based on microreactor and microfluidic technology is especially attractive for radiosyntheses involving short-lived positron emitters like fluorine-18 (t1/2=109.8 min). The technology has the potential that automated syntheses at the nanogram-tomicrogram scale can be extended to the synthesis of radiolabeled compounds. The advantages of micro-reactor
236
V.R. Bouvet et al. / Nuclear Medicine and Biology 38 (2011) 235–245
and microfluidic technology for PET tracer synthesis include the use of minute amounts of labeling precursor, reduced separation challenges and rapid reaction rates [8,9]. In this line, the micro-reactor-based multistep synthesis of 2-deoxy2-[18F]fluoro-D-glucose ([18F]FDG) was the first successful example of micro-reactor and microfluidic technology for the synthesis of a PET radiotracer [10]. Microfluidic technology was further applied to radiosyntheses of PET radiotracers involving iodine-124 and carbon-11 [11]. However, despite the numerous advantages of micro-reactor and microfluidic technology in organic synthesis, only few examples have been reported in the field of radiopharmaceutical sciences. Promising first results on the application of microfluidic technology have not yet led to the development of procedures for the routine production of radiopharmaceuticals such as [18F]FDG and [18F]FLT [12]. The clinical demand of [18F]FDG, [18F]FLT and other PET radiopharmaceuticals is increasing, and recent advances in radiotracer production should be combined with advantages of the micro-reactor and microfluidic technology [12]. Recently, we expanded the application of microfluidic technology to the radiosynthesis of various prosthetic groups like [18F]FDG-MHO and [18F]FBA, and their subsequent use in peptide synthesis using a NanoTek microfluidic synthesis system [13,14]. This led us to the conclusion that for full exploitation of microfluidic technology potential it is necessary to fully understand the influence of different reaction parameters on the performance of the microfluidic system and their impact on the radiochemical yield. Toward this goal, we elected to explore the radiosynthesis of PET radiotracer 1-(5-deoxy-5-fluoro-α-D-arabinofuranosyl)-2-nitroimidazole ([18F]FAZA) using the Advion NanoTek Microfluidic Synthesis System. [18 F]FAZA is an important radiotracer for imaging tumor hypoxia. Besides [18F]FAZA, several other hypoxia imaging radiotracers have been developed [15]. [18F]Fluoromisonidazole ([18F]MISO) is currently the most widely used clinical radiotracer for imaging tumor hypoxia [16,17]. [18F]FAZA was developed as an alternative azomycin-based imaging agent to improve image contrast by increasing the rate of blood clearance [18,19]. To date, [18F]FAZA has been used in several clinical trials [20,21] which require a reliable and stable radiopharmaceutical production of the radiotracer. However, current automated synthesis provides the radiotracer in widely varying radiochemical yields ranging from 5% to 25% [19,21]. In the present study, we have analyzed the complex array of different reaction conditions linked to the microfluidic synthesis system during the radiosynthesis of hypoxia imaging agent [18F]FAZA. Two different setup modes were used for the NanoTek Microfluidic Synthesis System to allow efficient radiosynthesis at low (1 to 50 MBq) and high (0.4 to 2.1 GBq) radioactivity levels. In addition, [18F]FAZA prepared by microfluidic technology was used for dynamic small animal PET studies in EMT-6 tumor-bearing mice.
2. Materials and methods 2.1. General All commercial reagents and solvents were used without further purification unless otherwise specified. Labeling precursor 1-(2,3-di-O-acetyl-5-O-tosyl-α-D-arabinofuranosyl)-2-nitroimidazole 1 and reference compound FAZA were obtained from ABX advanced biochemical compounds (Radeberg, Germany). Radiosyntheses were carried out in an Advion NanoTek Microfluidic Synthesis System (Ithaca, NY, USA) consisting of a microscale concentrator and evaporator module for [18F] fluoride drying, an injection module equipped with syringe pumps and distributions valves, and a micro-reactor module equipped with temperature controllers [22,23]. The unit is controlled by a standalone computer, which allows the setting of different parameters (flow rate, temperature and injected volumes). The reactors used in the experiments had a 100 μm inner diameter (ID) and length ranging from 2 m (16 μl) to 8 m (64 μl). HPLC purification of [18F]FAZA 3 was performed on a semi-preparative Nucleosil C18 column (100 Å, 10 μm, 250×10 mm). The eluting solvent was ethanol/water (5:95, v/v) at a flow rate of 4 ml/min. The radiochemical and chemical purity of intermediate 2 and [18F]FAZA 3 was determined using the same columns. The eluting solvent was 100% H2O for 5 min followed by a 10-min period from 0:100 (v/v) to 100:0 (v/v) acetonitrile/water. UV detection was performed at 220 and 320 nm. Radioactivity detection was performed using a wellscintillation NaI(Tl) detector. 2.2. Production of [18F]fluoride [18F]Fluoride was produced by the 18O(p,n)18F nuclear reaction through proton irradiation of enriched (98%) 18O water (3.0 ml, Rotem, Germany) using a TR19/9 cyclotron (Advanced Cyclotron Systems, Inc., Richmond, BC, Canada). 2.3. Drying of [18F]fluoride [18 F]Fluoride drying process was performed in the Advion NanoTek Microfluidic System concentrator module. The programmed sequence contained the following steps: cyclotron-produced [18F]fluoride in 2.5 ml of 18O-enriched H2O was passed through a Sep-Pak plus QMA cartridge. The cartridge was dried with air, and [18F]fluoride was eluted using 800 μl of K222/K2CO3 solution [kryptofix K222 (40 mg) in acetonitrile (1.7 ml); K2CO3 (10 mg) in water (0.5 ml)]. The solvent was evaporated to dryness, and the cartridge was eluted a second time with 800 μl of K222/ K2CO3 solution. The solution was subjected to three consecutive azeotropic distillations at 100°C using 500 μl of acetonitrile for each distillation step. [18F]Fluoride was redissolved in the desired solvent (e.g., acetonitrile, DMSO). The overall process required 19 min.
V.R. Bouvet et al. / Nuclear Medicine and Biology 38 (2011) 235–245 18
2.4. Synthesis of [ F]FAZA 18
Conventional synthesis of [ F]FAZA was performed according to literature [19]. Two different setup modes of the Advion NanoTek Microfluidic Synthesis System were used for the optimization at low radioactivity levels (1 to 50 MBq; Scheme 2) and upscale experiments at higher radioactivity levels (0.4 to 2.1 GBq; Scheme 3). Each reaction was repeated three times and the yields presented are the average of three reactions with their standard deviation (Tables 1 and 4). For the optimization of the reaction conditions, dried and redissolved [18F]fluoride and a solution of labeling precursor 1 were introduced into the system in 10 to 40 μl portions through dedicated injection loops. The [18F]fluoride radioactivity injected was recorded using a Capintec radioisotope calibrator just before its introduction into the loop system, which allowed for an accurate determination of the radioactivity injected for each reaction. Flow rate, reactor temperature and injected volumes were adjusted and controlled by the Advion NanoTek Microfluidic Synthesis unit software (v. 1.4) on a standalone computer. The percentage of available [18 F]fluoride radioactivity was determined by comparing the amount of radioactivity injected with the amount recovered out of the micro-reactors. A total volume of 400 μl of redissolved [18F]fluoride (0.4 to 2.1 GBq) in DMSO and up to 800 μl of labeling precursor 1 (2.5 or 5.0 mg/ml) in DMSO were used for the preparation of [18F]FAZA at higher radioactivity levels. The reaction mixture containing diacetylated intermediate 2 was transferred into a reaction vial with 100 μl of 0.1N NaOH solution at 25 °C and stirred for 2 min. The solution was neutralized with 200 μl of a 0.1N KH2PO4 solution, and the mixture was loaded onto semi-preparative HPLC for purification. The eluting solvent was ethanol/water (5:95, v/v) at a flow rate of 4 ml/min. The product eluting between 12.1 and 13.5 min was collected. The radiochemical and chemical purity of [18F]FAZA was determined using the same columns. The eluting solvent was 100% H2O for 5 min followed by a 10Table 1 Influence of the reaction temperature on the [18F]fluoride incorporationa Entry
Temperature (°C)
Volume ratiob
Residency time (min)c
Radiochemical yieldd (n=3) (%±S.D.)
1 2 3 4
80 100 120 140
2 2 2 2
3.55 3.55 3.55 3.55
b2 15±2 81±7 Degradation
a All reactions were carried out using an overall flow rate of 18 μl/min, a reactor size of 64 μl and a labeling precursor 1 concentration of 2.5 mg/ml. [18F]Fluoride bolus varied from 10 to 20 μl. Each reaction was repeated three times (n=3) and the yield presented is the average of the three reactions with their associated standard deviation. b Volume ratio=V[precursor]/V[18F]. c Residency time reflects the time the solution resided inside the reactor. d The radiochemical yields were determined by radio-HPLC referring to the percentage of the radioactivity area of 18F-labeled intermediate 2 to the total radioactivity area.
237
min period from 0:100 (v/v) to 100:0 (v/v) acetonitrile/water. The product was co-eluting with nonradioactive reference compound FAZA. Optimized reaction conditions at radioactivity levels of up to 2.1 GBq of dried and resolubilized [18F]fluoride included the use of labeling precursor 1 (5 mg/ml), a reactor size of 96 μl, an overall flow rate of 27 μl/min and a temperature of 120 °C. 2.5. Small animal PET study Positron emission tomography (PET) experiments were performed using EMT-6 tumor-bearing BALB/c mice. All animal experiments were carried out in accordance with the guidelines of the Canadian Council on Animal Care (CCAC) and under a protocol (AC 07137) approved by the local animal care committee (Cross Cancer Institute, University of Alberta). Murine EMT-6 cells (5×106 cells in 100 μl PBS) were injected into the upper right flank of female Balb/C mice (20–24 g, Charles River, Saint-Constant, Quebec, Canada). The EMT-6 tumor-bearing mice were imaged after allowing 7 to 10 days for tumors reaching sizes of about 300–500 mg. The mice were fasted for 3 to 4 h prior to the imaging experiments. The animals were anesthetized with isoflurane in 40% oxygen/60% nitrogen (gas flow, 1 L/min), and body temperature was kept constant at 37°C for the entire experiment. Mice were positioned and immobilized in the prone position with their medial axis parallel to the axial axis of the scanner and their thorax, abdomen and hind legs (organs of interest: heart, kidneys, bladder, tumors) in the centre of the field of view of the microPET R4 scanner (Siemens Preclinical Solutions, Knoxville, TN, USA). A transmission scan for attenuation correction was not acquired. The radioactivity of the injection solution in a 0.5-ml syringe was determined with a dose activimeter (Atomlab 300, Biodex Medical Systems, New York, USA). The emission scan of 90-min PET acquisition was started, and with a delay of approximately 15 s, 4–5 MBq of [18F] FAZA in 80–120 μl saline was injected through a needle catheter into a tail vein. Data acquisition continued for 90 min in 3D list mode. The list mode data was sorted into sinograms with 56 time frames (10×2, 8×5, 6×10, 6×20, 8×60, 10×120, 7×300 s). The frames were reconstructed using the ordered subset expectation maximization applied to 3D sinograms (3D OSEM). The pixel size was 0.085×0.085×0.12 cm, and the resolution in the centre field of view was 1.8 mm. No correction for partial volume effects was performed. The image files were further processed using the ROVER software v. 2.0.21 (ABX GmbH, Radeberg, Germany). Masks for defining threedimensional regions of interest (ROIs) were set and the ROIs were defined by thresholding. ROI time–activity curves were generated for subsequent data analysis. Standardized uptake values [SUV=(activity/ml tissue)/(injected activity/ body weight), ml/g] were calculated for each ROI.
238
V.R. Bouvet et al. / Nuclear Medicine and Biology 38 (2011) 235–245
3. Results and discussion 3.1. Optimization of reaction parameters at low radioactivity levels (1 to 50 MBq) The synthesis of [18F]FAZA 3 involves a two-step process consisting of (1) incorporation of [18F]fluoride into tosylate labeling precursor 1 via nucleophilic radiofluorination followed by (2) removal of the acetyl protecting groups in intermediate 2 using basic hydrolysis with 0.1N NaOH at room temperature. The reaction sequence for the radiosynthesis of [18F]FAZA is depicted in Fig. 1. Cyclotron-produced [18F]fluoride was converted into powerful nucleophilic radiofluorination agent [18F]KF by means of kryptofix K222 and K2CO3 in acetonitrile/water and subsequent azeotropic distillation using the Advion NanoTek concentrator as displayed in Scheme 1. Prior to the synthesis of [18F]FAZA employing microfluidic technology, we carried out the radiosynthesis according to the method described by Reischl et al. [19]. Optimum reaction conditions (5 mg/ml of labeling precursor 1, 20 min at 120°C) afforded 18F-labeled intermediate 2 in up to 62% radiochemical yield. This is in agreement with the results previously reported by Reischl et al. [19]. Two different setup modes of the Advion NanoTek Microfluidic Synthesis System were used for [18F]fluoride incorporation into labeling precursor 1 as the first step within the radiosynthesis sequence of [18F]FAZA. In the first set of reactions, reaction conditions for [18F]fluoride incorporation were optimized at low radioactivity levels (1–50 MBq) by varying different reaction parameters (reaction temperature, flow rate, residence time, concentration of labeling precursor 1, applied volume ratio between labeling precursor 1 and [18F]fluoride) using a microfluidic reactor setup as shown in Scheme 2. A second setup mode of the Advion NanoTek Microfluidic Synthesis System (Scheme 3) was used for the synthesis of [18F]FAZA starting with higher radioactivity levels of [18F]fluoride (0.4 to 2.1 GBq). In the first experimental setup of the microfluidic reactor system (Scheme 2), labeling precursor 1 and redissolved [18F]KF were both introduced into the system through their dedicated injection loops. Once the reactants (labeling precursor 1 and resolubilized [18F]KF) were transferred onto the injection loops, the solvent (DMSO) was pushed through Port E to deliver the reactants into the microreactor through Port C (right panel of Scheme 2). In the first set of reactions, we tested the influence of the reaction temperature on the radiochemical
Fig. 1. Radiosynthesis of [18F]FAZA.
Scheme 1. The Advion NanoTek concentrator.
yield as reflected by the incorporation of [18F]fluoride into labeling precursor 2. The results are summarized in Table 1. Results in Table 1 show that in the case of microfluidic synthesis the optimum reaction temperature for the incorporation of [18 F]fluoride into labeling precursor 1 via nucleophilic radiofluorination is 120°C. At lower temperatures, incorporation yield of [18F]fluoride drops significantly to 15±2% at 100°C (Entry 3) and smaller than 2% at 80°C (Entry 1). At higher reaction temperature of 140°C (Entry 4), substantial degradation occurred which resulted in the formation of various radioactive products containing only a small amount of desired compound 2. Moreover, the complex reaction mixture also complicates subsequent isolation and purification of [18F]FAZA after hydrolysis by means of HPLC. Performance of the radiosynthesis at 120°C employing conventional reaction conditions also led to some degradation. Direct comparison of radiochemical yields obtained with conventional reaction vs. microfluidic technology while employing comparable labeling precursor concentration (2.5 mg/ml) clearly demonstrates the advantage of microfluidic technology. After a residency time of 3.55 min at 120°C, application of microfluidic technology provided a radiochemical yield of 81±7% (Entry 3, Table 1), whereas conventional synthesis afforded only 27%. Moreover, reactions using microfluidic technology were carried out with very low labeling precursor 1 amounts (38 μg) for each synthesis. In a second set of reactions, we explored the influence of the flow rate on the incorporation of [18F]fluoride into labeling precursor 1. The flow rate at which the reaction mixtures are entering the reactor is directly linked with two additional reaction parameters: the mixing of the reactants and the residence time of the aliquot within the reactor. In order to change the flow rate of the reactions while keeping all the other parameters constant, especially the residence time within the reactor, it was necessary to perform the
V.R. Bouvet et al. / Nuclear Medicine and Biology 38 (2011) 235–245
239
Scheme 2. Microfluidic reactor setup mode to optimize reaction conditions for [18F]fluoride incorporation into labeling precursor 1 at low radioactivity levels.
reactions in different reactors with different sizes. For this purpose, [18F]fluoride incorporation into labeling precursor 1 was carried out at 120°C in one 32-μl reactor, one 64-μl reactor or two 64-μl reactors linked in series, respectively. Table 2 summarizes the influence of different flow rates on the radiochemical yield. Fig. 2 shows various radio-HPLC profiles within the synthesis sequence of [18F]FAZA using optimal reaction conditions. Data of Entries 1, 2 and 3 in Table 2 clearly demonstrate the positive effect of an increased flow rate on the radiochemical yield while maintaining the reactor residency time constant at 3.55 min. High radiochemical yields of up to 86±7% were obtained in a 128-μl reactor (two 64-μl reactors) at an overall flow rate of 36 μl/min (Entry 3). Higher flow rates seem to provide higher radiochemical yields as exemplified by the experiments outlined in Entries 4 and 5. A possible explanation for this observation could be the more efficient mixing of both reactants within the reactor at higher flow rates [24]. Such a chaotic mixing would be directly linked to the viscosity of the solvent and the inner diameter of the reactor. Therefore, it seems reasonable to conclude that each solvent possesses its own optimum flow rate for a reactor with a given inner diameter. Our recent peptide labeling experiments using acetonitrile, water, DMF and DMSO as solvents seem to indicate that this phenomenon is more predominant as the viscosity of the solvent increases. However, steady increase of the flow rate is limited. Higher flow rates require the usage of larger reactors to provide sufficient residency times for the reaction. On the other hand, larger reactors impose higher back pressure on the overall pump system, which may lead to a shut-off of the
system. Moreover, incorporation of [18 F]fluoride into labeling precursor 2 proceeds best with DMSO as the solvent. DMSO is a highly viscous solvent which causes substantial back pressure. Performance of the reaction in two 64-μl reactors linked in series at an overall flow rate of 36 μl/min (Entry 3) resulted in the maximum acceptable back pressure of 400 psi as recommended by the manufacturer for the operation of the NanoTek Microfluidic Synthesis System. Another reaction parameter that can be controlled by the microfluidic system is the volume ratio (Vprecursor/V18 [ F]) between labeling precursor 1 and [18F]fluoride injected into the reactor. The influence of the volume ratio on the radiochemical yield is given in Table 3. The volume ratios in this set of reactions were 1 (Entry 1), 2 (Entry 2) and 3 (Entry 3). Best radiochemical yields were Table 2 Influence of the flow rate on the [18F]fluoride incorporationa Entry
Volume ratiob
Overall flow rate (μl/min)
Reactor size (μl)
Residency time (min)c
Radiochemical yieldd (n=3) (%±S.D.)
1 2 3 4 5
2 2 2 1 1
9 18 36 12 18
32 64 128 64 64
3.55 3.55 3.55 5.33 3.55
59±4 81±7 86±7 64±3 71±2
a All reactions were carried out using a reaction temperature of 120°C and a labeling precursor 1 concentration of 2.5 mg/ml. [18F]Fluoride bolus varied from 15 to 40 μl. b Volume ratio=V[precursor]/V18 [ F]. c Residency time reflects the time the solution resided inside the reactor. d The radiochemical yields were determined by radio-HPLC referring to the percentage of the radioactivity area of 18F-labeled intermediate 2 to the total radioactivity area.
240
V.R. Bouvet et al. / Nuclear Medicine and Biology 38 (2011) 235–245
Fig. 2. Radio-HPLC and UV (320 nm) profiles: (A) Synthesis of intermediate 2 using optimal reaction conditions (Entry 3, Table 2). (B) Deprotection of intermediate 2. No HPLC purifications were carried out between [18F]fluoride incorporation and the deprotection step. (C) [18F]FAZA after HPLC purification, concentration and saline buffer solubilization. Radio-HPLC profiles using optimal reaction conditions.
obtained by using a volume ratio of 2 (81±7%) or 3 (86±1%). Application of a volume ratio of 1 gave slightly lower radiochemical yields of 71±4% (Entry 1). The different volume ratios were achieved by injecting the respective volume equivalents of the labeling precursor 1 into the mixing chamber of the reactor. To achieve a volume ratio of 2 (Entry 2), two volume equivalents (20 μl) of labeling Table 3 Influence of the injected volume ratio (Vprecursor/V18 [ F]) between the labeling precursor 1 and [18F]fluoride on the radiochemical yielda Entry
Volume ratiob
Overall flow rate (μl/min)
Residency time (min)c
Radiochemical yieldd (n=3) (%±S.D.)
1 2 3
1 2 3
18 18 20
3.55 3.55 3.20
71±4 81±7 86±1
a All reactions were carried out using a reaction temperature of 120 °C, a labeling precursor 1 concentration of 2.5 mg/ml and a reactor size of 64 μl. [18F]Fluoride bolus varied from 10 to 40 μl. b Volume ratio=V[precursor]/V18 [ F]. c Residency time reflects the time the solution resided inside the reactor. d The radiochemical yields were determined by radio-HPLC referring to the percentage of the radioactivity area of 18F-labeled intermediate 2 to the total radioactivity area.
precursor 1 were injected at a flow rate of 12 μl/min for each volume equivalent (10 μl at a flow rate of 6 μl/min) of [18F]fluoride into the mixing chamber of the reactor. This resulted in an overall flow rate of 18 μl/min. Increase of the volume ratio from 2 to 3 seemed not to result in significantly higher radiochemical yields (81±7% for Entry 2 vs. 86±1% for Entry 3). Consequently, further increase of the volume ratio from 2 to 3 is not useful. The most efficient volume 18 ratio Vprecursor/V18 [ F] for the incorporation of [ F]fluoride into labeling precursor 1 was 2. The influence of the reactor size, which is directly linked with the residency time, was studied in another set of reactions while all other reaction conditions remained constant (labeling precursor concentration of 2.5 mg/ml, reaction temperature of 120 °C, volume ratio of 2 and an overall flow rate of 18 μl/min). Depending on the reactor size, Table 4 shows the results for different residence times of the reaction mixture within the reactor. Maximum radiochemical yields of 86±4% were obtained when a 64-μl reactor was linked in series to a 32-μl reactor to give a total reactor size of 96 μl (Entry 3). With the use of this combination of reactors, the reaction was completed within 5.33 min. A reactor size of 64 μl and a corresponding
V.R. Bouvet et al. / Nuclear Medicine and Biology 38 (2011) 235–245 Table 4 Influence of the residency time on the [18F]fluoride incorporationa Entry
Reactor size (μl)
Residency time (min)b
Radiochemical yieldc (n=3) (%±S.D.)
1 2 3 4
32 64 96 128
1.78 3.55 5.33 7.12
38±5 81±7 86±4 34±2 +degradation
a All reactions were carried out using a reaction temperature of 120°C, a labeling precursor 1 concentration of 2.5 mg/ml, a volume ratio of 2 and an overall flow rate of 18 μl/min. [18F]Fluoride bolus varied from 15 to 20 μl. b Residency time reflects the time the solution resided inside the reactor. c The radiochemical yields were determined by radio-HPLC referring to the percentage of the radioactivity area of 18F-labeled intermediate 2 to the total radioactivity area.
residency time of 3.55 min gave slightly lower radiochemical yields of 81±7 (Entry 2). A substantially smaller reactor of 32 μl with a short residence time of 1.78 min resulted in significantly lower radiochemical yields of 38±5% (Entry 1). Extension of the reactor size to 128 μl (residency time of 7.12 min) resulted in a comparably moderate 34±2% radiochemical yield (Entry 4). The reaction mixture contained many unidentified radioactive degradation products similar to the ones discussed in Table 1 (Entry 4). They are most likely due to the prolonged residency time of 7.12 min at elevated temperatures of 120°C. It is known from many radiosyntheses involving the short-lived positron emitter fluorine-18 that the obtained radiochemical yield directly correlates to the amount of labeling precursor. The important effect of labeling precursor amount on the radiochemical yield has been observed for various direct nucleophilic radiolabeling reactions as well as for indirect radiolabeling reactions involving 18F-labeled prosthetic groups. In the last set of reaction, we studied the influence of different labeling precursor concentrations on the radiochemical yield of [18F]FAZA. The results are summarized in Table 5. The results displayed in Table 5 clearly demonstrate the importance of the amount of labeling precursor on the radiochemical yield. Only very little product (b5%) was formed when low concentration (1 mg/ml) of labeling precursor 1 was used (Entry 1). Significantly higher Table 5 Influence of the precursor concentration on the [18F]fluoride incorporationa Entry
Labeling precursor 1 concentration (mg/ml)
Radiochemical yieldb (n=1) (%)
1 2 3
1.0 2.5 5.0
b5 81 87
a All reactions were carried out using a reaction temperature of 120°C, a volume ratio of 2, an overall flow rate of 18 μl/min and a reactor size of 64 μl (residency time of 3.55 min). [18F]Fluoride bolus varied from 15 to 20 μl. b The radiochemical yields were determined by radio-HPLC referring to the percentage of the radioactivity area of 18F-labeled intermediate 2 to the total radioactivity area.
241
radiochemical yields of 81% and 87% were obtained at higher concentrations of 2.5 mg/ml (Entry 2) and 5.0 mg/ml (Entry 3). This finding is consistent with results reported in the literature emphasizing the importance of labeling precursor amount upon the radiochemical yield [19,25]. The obtained high radiochemical yields of 81% and 87% as shown in Entries 2 and 3 of Table 5 further confirm the superior labeling efficiency of microfluidic technology compared to conventional synthesis when similar reaction conditions are applied [19]. During the optimization process (1 to 50 MBq radioactivity level) for the incorporation of [18F]fluoride into [18F] FAZA labeling precursor 1, we used a setup mode of the NanoTek Microfluidic Synthesis System displayed in Scheme 2. Best results were obtained using the following reaction conditions: 50 μg of labeling precursor 1 in 20 μl of DMSO (2.5 mg/ml), overall flow rate=18 μl/min, volume ratio (Vprecursor/V18 [ F])=2, reaction temperature=120°C, residence time=3.55 min. These parameters afforded radiochemical yields of greater than 80% for the first step of the reaction sequence within the synthesis sequence of [18F]FAZA. This represents a significant improvement compared to conventional radiolabeling method especially with respect to the amount of labeling precursor used and reaction [19]. Application of optimized reaction conditions in combination with subsequent basic hydrolysis of intermediate 2 with 0.1N NaOH at room temperature afforded [18F]FAZA in 63% radiochemical yield (decay-corrected) including HPLC purification. Hydrolysis was achieved by transferring the reaction mixture containing intermediate 2 into a 1-ml vial containing 0.1N NaOH. Hydrolysis of intermediate 2 occurred immediately and completely at room temperature. The total reaction time excluding HPLC purification was 8.78 min (1.67 min for the injection of 30 μl reaction mixture, 5.33 min residence time in the reactor, 1.78 min to transfer the reaction mixture from the reactor into the hydrolysis vial). 3.2. Radiosynthesis of [18F]FAZA at higher radioactivity levels (0.5 to 2.1 GBq) Application of optimized reaction conditions in the NanoTek Microfluidic Synthesis System setup mode as depicted in Scheme 2 resulted in unexpected problems when higher radioactivity levels (N200 MBq) were used. Substantial amounts of radioactivity (N75%) were trapped in the [18F]fluoride injection loop and other tubing of the apparatus. This phenomenon was only observed when DMSO was used as the solvent. Other solvents like acetonitrile did not show this effect. However, radiochemical yields for [18F]fluoride incorporation into labeling precursor 1 were much lower (b40%) in acetonitrile compared to DMSO as the solvent. Table 6 summarizes the obtained radiochemical yields for the synthesis of [18F]FAZA in DMSO using setup mode according to Scheme 2 and higher radioactivity levels (N200 MBq).
242
V.R. Bouvet et al. / Nuclear Medicine and Biology 38 (2011) 235–245
Table 6 Radiosynthesis of [18F]FAZA at higher radioactivity levels using original setup modea Entry
Radioactivity (MBq)
Incorporation yield (%)b
Available 18F radioactivity (%)c
Isolated radiochemical yieldd of [18F]FAZA (n=1) (%)
1 2 3
342 414 210
75 95 78
21 18 24
14 15 17
a All reactions were carried out using a reaction temperature of 120°C, a volume ratio of 2, an overall flow rate of 18 μl/min, a labeling precursor 1 concentration of 2.5 mg/ml and a reactor size of 64 μl (residency time of 3.55 min). [18F]Fluoride bolus ∼400 μl. b The radiochemical yields were determined by radio-HPLC referring to the percentage of the radioactivity area of 18F-labeled intermediate 2 to the total radioactivity area. c Available 18F radioactivity was determined by counting 18F radioactivity prior to and after passage through the injection loop and reactor. d After HPLC purification, decay-corrected.
Results in Table 6 show that incorporation of [18F] fluoride into the labeling precursor at higher radioactivity levels proceeded with comparable efficiency as found with lower radioactivity levels (Table 5, Entry 2). However, significant differences were found for the isolated radiochemical yields of [18F]FAZA at low radioactivity level (63%) and higher radioactivity level (15%). As shown in Table 6, more than 75% of starting [18F]fluoride radioactivity was not available for the radiosynthesis at higher radioactivity levels. Setup mode of the NanoTek Microfluidic Synthesis System as used for the optimization of the reaction conditions at low radioactivity levels according to Scheme 2 seems to be not suitable for the application of higher radioactivity amounts. As a consequence, we
Scheme 3. Microfluidic reactor setup mode for the radiosynthesis of [18F] FAZA at higher radioactivity levels.
modified the microfluidic system to avoid the observed losses of [18F]fluoride radioactivity in the [18F]fluoride injection loop and connected tubing. The modified setup mode is given in Scheme 3. In the modified setup mode, dried [18F]fluoride in DMSO and labeling precursor 1 in DMSO are introduced into the micro-reactor while omitting the injection loops. Employing this procedure provided significantly improved available 18F radioactivity of more than 70%. The results of [18F]FAZA synthesis in the modified setup mode at higher radioactivity levels are summarized in Table 7. Optimized reaction conditions at radioactivity levels of up to 2.1 GBq of dried and redissolved [18F]fluoride included the use of a labeling precursor 1 concentration of 5 mg/ml and a reactor size of 96 μl. Based on an overall flow rate of 27 μl/min the residency time in the 96-μl reactor was 3.55 min to yield 40% (decay-corrected) of radiochemically pure [18F]FAZA after removal of the protection groups and HPLC purification (Entry 4). The total synthesis time for [18F] FAZA was 55 to 60 min including HPLC purification. The specific activity was determined to be in the range of 70 to 150 GBq/μmol at the end of synthesis. Lower labeling precursor amounts (2.5 mg/ml, Entry 2) gave lower radiochemical yields of 25%. A slower overall flow rate of 18 μl/min, in order to prolong the residence time of the reaction partners in the reactor, did not increase the radiochemical yield (Entries 1, 3 and 5). The isolated radiochemical yield of 40% for [18F]FAZA using microfluidic technology represents a significant improvement to the radiochemical yields of 5% to 25% reported in the literature based on conventional synthesis. 3.3. Small animal PET study [18F]FAZA prepared via microfluidic technology was used for tumor uptake studies in EMT-6 tumor-bearing BALB/c mice. The murine mammary breast cancer model EMT-6 was used in previous radiopharmacological studies to analyze uptake of various radiotracers for measuring hypoxia [26,27]. Fig. 3 shows registered small animal PET images at different time points during a dynamic PET study of 90 min. During the first minutes after radiotracer administration, the radiotracer was rapidly distributed into different body compartments. Radioactivity uptake in the EMT-6 tumor was visible after 30 min pi. Time–activity curves derived from a properly defined ROIs over the heart (blood pool), muscle and tumor were generated from four individual small animal PET experiments (Fig. 4). Blood clearance of 18F radioactivity throughout the body was fast. After 15 min (900 s), less than 10% of the maximum blood uptake (ROI over the heart) was detectable: SUVmax 14.80±5.49 vs. SUV15 min 1.39±0.23, respectively. After 90 min pi, only 4% of the maximum blood uptake was observed. These observations agree with previous biodistribution studies which have shown fast blood clearance of
V.R. Bouvet et al. / Nuclear Medicine and Biology 38 (2011) 235–245
243
Table 7 Radiosynthesis of [18F]FAZA at higher radioactivity levels using modified setup modea Entry
Radioactivity (MBq)
Overall flow rate (μl/min)
Residency time (min)
Incorporation yield (%)b
Available 18F radioactivity (%)c
RCYd [18F]FAZA (n=1) (%)
1 2 3 4 5
1257 1110 856 2104 398
18 27 18 27 18
5.33 3.55 5.33 3.55 7.11
23 32 38 47±5 25
85 85 75 90±2 70
18e,c 25e,g 27f,g 40±3f,h 16f,i
All reactions were carried out using a reaction temperature of 120°C and a volume ratio of 2. [18F]Fluoride bolus ∼400 μl. The radiochemical yields were determined by radio-HPLC referring to the percentage of the radioactivity area of 18F-labeled intermediate 2 to the total radioactivity area. c Available 18F radioactivity was determined by counting 18F radioactivity prior to and after passage through the injection loop and reactor. d RCY=Radiochemical yield, after HPLC purification, decay-corrected. e Labeling precursor 1 concentration of 2.5 mg/ml. f Labeling precursor 1 concentration of 5.0 mg/ml. g Reactor size of 96 μl (32 μl+64 μl). h Entry 4 was carried out three times. i Reactor size of 128 μl (64 μl+64 μl). a
b
[18F]FAZA in different tumor-bearing mouse models [26]. However, a recent dynamic small animal PET study using [18F]FAZA in SCCVII tumor-bearing C3H/HeNTaC mice showed slower blood clearance reaching 14–21% of the maximum blood uptake after 90 min pi. This finding may be attributed to the different mouse strain used [28]. Radioactivity uptake in EMT-6 tumors reached a maximum after 20 min (1200 s) with a SUV20 min of 0.74±0.08. Over the time course of the dynamic PET experiments, tumor uptake slightly decreased reaching a SUV90 min of 0.62±0.12 (n=4)
after 90 min pi. However, in muscle as reference tissue, the maximum SUV was comparable to that of the tumor after 20 min pi (SUV20 min of 0.73±0.16). Radioactivity washout from muscle was fast reaching a SUV90 min of 0.44±0.14 (n=4; Pb.01) after 90 min pi. Based on the muscle and blood clearance profiles of 18F radioactivity, the tumor/muscle and tumor/blood ratio increased continuously over time. Increased tumor/muscle and tumor/blood ratios are characteristic parameters of a radiotracer for imaging tumor hypoxia with PET [26–29].
Fig. 3. Coronal small animal PET images of EMT-6 tumor-bearing mice after injection of [18F]FAZA (5 MBq) into the tail vain. Isoflurane (1.5%) was used for anesthesia.
244
V.R. Bouvet et al. / Nuclear Medicine and Biology 38 (2011) 235–245
Fig. 4. Representative time–activity curves of 18F radioactivity concentration and tumor/muscle and tumor/blood ratio during the entire study period of 90 min after the injection of [18F]FAZA (5 MBq).
4. Conclusion We have investigated the influence of different reaction parameters of the NanoTek Microfluidic Synthesis System exemplified by the optimization of the radiosynthesis of hypoxia imaging agent [18F]FAZA. Careful screening of various reaction parameters like labeling precursor amount, reaction temperature, flow rate and reactor size provided a deeper understanding of how these reaction parameters influence the performance of the microfluidic device to achieve reasonable radiochemical yields for radiofluorinations starting from [18F]fluoride. It was shown that depending on the radioactivity level, two different setup modes for the microfluidic system are necessary to achieve reasonable radiochemical yields at low and moderate to high radioactivity levels. The obtained radiochemical yields were sufficient for subsequent preclinical radiopharmacological evaluations as shown for the dynamic small animal PET studies. The modified setup mode for higher radioactivity levels would allow the preparation of sufficient amounts of [18F]FAZA as a single dose for patient studies. The obtained results clearly demonstrate the potential of microfluidic technology in radiopharmaceutical chemistry for the fast and reliable synthesis of radiotracers while using only small amounts of labeling precursor. The achieved high incorporation yields (N85%) of [18F]fluoride into labeling precursor 1 via nucleophilic radiofluorination and the possibility to connect several micro-reactors in series allow performance of complex multistep radiosynthesis in the microfluidic synthesis system. Moreover, the advantage of very fast reaction times in the 2- to 4-min range allows
application of radiotracer synthesis involving other shortlived positron emitters like carbon-11 (t1/2=20.4 min) and nitrogen-13 (t1/2=9.9 min). In conclusion, microfluidic technology is a viable approach for the rapid and reliable synthesis of radiolabeled compound. Promising results for the synthesis of [18F]FAZA warrant extension of microfluidic technology to the synthesis of other clinically relevant PET radiotracers. Acknowledgment The authors would like to thank John Wilson, David Clendening and Jayden Sader from the Edmonton PET Center for radionuclide production and excellent technical support. F.W. thanks the Dianne and Irving Kipnes Foundation for supporting this work. References [1] DeWitt SH. Microreactors for chemical synthesis. Curr Opin Chem Biol 1999;3:350–6. [2] Fletcher PDI, Haswell SJ, Pombo-Villar E, Warrington BH, Watts P, Wong SYF, et al. Micro reactors: principles and applications in organic synthesis. Tetrahedron 2002;58:4735–57. [3] Haswell SJ, Watts P. Green chemistry: synthesis in micro reactors. Green Chem 2003;5:240–9. [4] Jähnisch K, Hessel V, Löwe H, Baerns M. Chemistry in microstructured reactors. Angew Chem Int Edit 2004;43:406–46. [5] Feng XZ, Haswell SJ, Watts P. Organic synthesis in micro reactors. Curr Top Med Chem 2004;4:707–27. [6] Cullen CJ, Wootton RC, de Mello AJ. Microfluidic systems for highthroughput and combinatorial chemistry. Curr Opin Drug Discov Devel 2004;7:798–806.
V.R. Bouvet et al. / Nuclear Medicine and Biology 38 (2011) 235–245 [7] Yoshida JI. Flash chemistry using electrochemical method and microsystems. Chem Commun 2005;36:4509–16. [8] Lu SY, Watts P, Chin FT, Hong J, Musachio JL, Briard E, et al. Syntheses of C-11- and F-18-labeled carboxylic esters within a hydrodynamically-driven micro-reactor. LabChip 2004;4:523–5. [9] Miller PW, Long NJ, de Mello AJ, Vilar R, Audrain H, Bender D, et al. Rapid multiphase carbonylation reactions by using a microtube reactor: applications in positron emission tomography 11C-radiolabeling. Angew Chem Int Edit 2007;46:2875–8. [10] Lee CC, Sui GD, Elizarov A, Shu CYJ, Shin YS, Dooley AN, et al. Multistep synthesis of a radiolabeled imaging probe using integrated microfluidics. Science 2005;310:1793–6. [11] Gillies JM, Prenant C, Chimon GN, Smethurst GJ, Perrie W, Hamblett I, et al. Microfluidic reactor for the radiosynthesis of PET radiotracers. Appl Radiat Isot 2006;64:325–32. [12] Elizarov AM, van Dam RM, Shin YS, Kolb HC, Padgett HC, Stout D, et al. Design and optimization of coin-shaped microreactor chips for PET radiopharmaceutical synthesis. J Nucl Med 2009;51:282–7. [13] Wuest F, Berndt M, Bergmann R, Pietzsch J. Synthesis and application of [18F]FDG-maleimidehexyloxime ([18F]FDG-MHO): a [18F]FDGbased prosthetic group for the chemoselective 18F-labeling of peptides and proteins. Bioconjugate Chem 2008;19:1202–10. [14] Iwata R, Pascali C, Bogni A, Horvath G, Kovacs Z, Yanai K, et al. A new, convenient method for the preparation of 4-[18F]fluorobenzyl halides. Appl Radiat Isot 2000;52:87–92. [15] Rajendran JG, Krohn KA. Imaging hypoxia and angiogenesis in tumors. Radiol Clin North Am 2005;43:169–87. [16] Chapman JD, Engelhardt EL, Stobbe CC, Schneider RF, Hanks GE. Measuring hypoxia and predicting tumor radioresistance with nuclear medicine assays. Radiother Oncol 2008;46:229–37. [17] Rajendran JG, Mankoff DA, O'Sullivan F, Peterson LM, Schwartz DL, Conrad EU, et al. Hypoxia and glucose metabolism in malignant tumors: evaluation by [18F]fluoromisonidazole and [18F]fluorodeoxyglucose positron emission tomography imaging. Clin Cancer Res 2004;10:2245–52. [18] Kumar P, Stypinski D, Xia H, McEwan AJB, Machulla HJ, Wiebe LI. Fluoroazomycin arabinoside (FAZA): synthesis, 2H and 3H-labelling and preliminary biological evaluation of a novel 2-nitroimidazole marker of tissue hypoxia. J Label Compds Radiopharm 1999;42:3–16.
245
[19] Reischl G, Ehrlichmann W, Bieg C, Solbach C, Kumar P, Wiebe LI, et al. Preparation of the hypoxia imaging PET tracer [18F]FAZA: reaction parameters and automation. Appl Radiat Isot 2005;62: 897–901. [20] Souvatzoglou M, Grosu AL, Röper B, et al. Tumour hypoxia imaging with [18F]FAZA PET in head and neck cancer patients: a pilot study. Eur J Nucl Med Mol Imaging 2007;34:1566–75. [21] Postema EJ, McEwan AJ, Riauka TA, Kumar P, Richmond DA, Abrams DN, et al. Initial results of hypoxia imaging using 1-alpha-D(5-deoxy-5-[ 18 F]-fluoroarabinofuranosyl)-2-nitroimidazole ( 18 FFAZA). Eur J Nucl Med Mol Imaging 2009;10:1565–73. [22] Briard E, Zoghbi SS, Siméon FG, Imaizumi M, Gourley JP, Shetty HU, et al. Single-step high-yield radiosynthesis and evaluation of a sensitive 18F-labeled ligand for imaging brain peripheral benzodiazepine receptors with PET. J Med Chem 2009;52:688–99. [23] Chun JH, Lu S, Lee YS, Pike VW. Fast and high-yield microreactor syntheses of ortho-substituted [18F]fluoroarenes from reactions of [18F] fluoride ion with diaryliodonium salts. J Org Chem 2010;75:3332–8. [24] Bringer MR, Gerdts CJ, Song H, Joshua D, Ismagilov T, Ismagilov RF. Microfluidic systems for chemical kinetics that rely on chaotic mixing in droplets. Phil Trans R Soc Lond A 2004;362:1087–104. [25] Wuest F, Hultsch C, Bergmann R, Johannsen B, Henle T. Radiolabelling of isopeptide N epsilon-(gamma-glutamyl)-L-lysine by conjugation with N-succinimidyl-4-[18F]fluorobenzoate. Appl Radiat Isot 2003;59:43–8. [26] Piert M, Machulla HJ, Picchio M, et al. Hypoxia-specific tumor imaging with F-18-fluoroazomycin arabinoside. J Nucl Med 2005;46: 106–13. [27] Beck R, Roper B, Carlsen JM, Huisman MC, Lebschi JA, Andratschke N, et al. Pretreatment F-18-FAZA PET predicts success of hypoxiadirected radiochemotherapy using tirapazamine. J Nucl Med 2007;48: 973–80. [28] Busk M, Horsman MR, Jakobsen S. Can hypoxia-PET map hypoxic cell density heterogeneity accurately in an animal tumor model at a clinically obtainable image contrast? Radiother Oncol 2009;92:429–36. [29] Reischl G, Dorow DS, Cullinane C, Katsifis A, Roselt P, Binns D, et al. Imaging of tumor hypoxia with [I-124] IAZA in comparison with [F-18]FMISO and [F-18]FAZA — first small animal PET results. J Pharm Sci 2007;10:203–11.
