Research Article Received 17 June 2013,
Revised 22 July 2013,
Accepted 18 August 2013
Published online 23 September 2013 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jlcr.3115
Development of a new precursor-minimizing base control method and its application for the automated synthesis and SPE purification of [18F]fluoromisonidazole ([18F]FMISO) Sang Ju Lee,a Ji Suk Hyun,a,b Seung Jun Oh,a* Kook Hyun Yu,b and Jin Sook Ryua Bases such as potassium carbonate and potassium bicarbonate (KHCO3) are essential for the elution of trapped [18F]fluoride from ion exchange cartridges and for the prevention of [18F]fluoride adsorption on the silica glass vial during the preparation of radiopharmaceuticals for positron emission tomography imaging. However, these bases promote the chemical decomposition of precursor compounds and the creation of unwanted organic impurities. Thus, the goal of the current study was to develop a new method for synthesizing [18F]fluoride-labeled radiopharmaceuticals (e.g., [18F] fluoromisonizadole ([18F]FMISO)) that permits the fine control of base concentrations while minimizing adverse events. Non-decay-corrected radiochemical yields of 25.1 ± 5.0% and 13.3 ± 5.1% (n = 3) were achieved after solid-phase extraction purification using automatic synthesis with GE TRACERlab MX and KHCO3 at concentrations of 14.1 and 33.0 μmol, respectively, and 1 mg of precursor (1-(2′-nitro-1′-imidazolyl)-2-O-tetra-hydropyranyl-3-O-toluenesulfonyl propanediol (NITTP)). The newly developed synthesis protocol with fine base control and low precursor content permits high radiochemical yields with minimal impurities. Keywords: [18F]fluormisonizadole; base control; [18F]fluorination; solid-phase extraction; automated synthesis; positron emission tomography
Introduction
J. Label Compd. Radiopharm 2013, 56 731–735
a Department of Nuclear Medicine, Asan Medical Center, University of Ulsan College of Medicine, 388-1 Pungnap-dong, Songpa-gu, Seoul 138-736, Korea b
Department of Chemistry, Dong Guk University, Seoul, Korea
*Correspondence to: Seung Jun Oh, Department of Nuclear Medicine, Asan Medical Center, 388-1 Pungnap-dong, Songpa-gu, Seoul 138-736, South Korea. E-mail: [email protected]
Copyright © 2013 John Wiley & Sons, Ltd.
731
Efficient [18F]fluorination of radiopharmaceuticals for positron emission tomography (PET) imaging generally requires high reaction temperatures and amount of precursor, as well as a long reaction time with [18F]fluoride, an appropriate solvent, and the catalyst. Potassium carbonate (K2CO3), potassium bicarbonate (KHCO3), and other bases are also essential for the efficient promotion of [18F]fluorination reactions because of their ability to elute trapped [18F]fluoride from ion exchange cartridges.1–3 However, the inclusion of such bases in [18F]fluorination procedures can lead to unwanted side reactions, including the decomposition of precursor, the concomitant formation of organic impurities, and necessitating a cumbersome purification step (e.g., highperformance liquid chromatography (HPLC) separation) to remove the organic contaminants. Furthermore, precursor decomposition is a common cause of low [18F]fluorinated product yields.4 The present study utilized precise base control in an attempt to overcome the complications of organic impurities contamination and low product yield. Precise control of base concentrations during [18F]fluorination procedures has a number of advantages, including the capacity to use minimal amounts of the precursor and the potential to readily purify [18F]fluorinated substrates along with a decreased occurrence of side reactions. For example, several investigations reported that control of tetrabutylammonium hydroxide (TBAOH) base levels in radiochemical synthesis reactions
reduced precursor requirements and the incidence of side reactions, while at the same time facilitating a high yield of [18F] fluorination.4–7 Nonetheless, a direct addition of TBAOH to the reactor prior to [18F]fluoride elution failed to demonstrate high radiochemical yields, the minimization of organic impurities, and straightforward application to automated synthesis systems; this was because TBAOH added in this manner could not effectively elute [18F]fluoride from the ion exchange cartridge.4 We previously developed a novel protocol to elute [18F]fluoride with a neutral salt solution and applied this methodology to the automated synthesis of [18F]fluoropropyl-carbomethoxyiodophenyl-nor-tropane ([18F]FP-CIT), a radioactive tracer.7 We hypothesized that a salt-based elution of trapped [18F]fluoride would allow for the additional direct application of base to the reactor and the fine control of base concentrations. According to this hypothesis, [18F]fluoride was eluted with a potassium methanesulfonate (KOMs) solution, and a precise amount of
S. J. Lee et al. TBAOH was then added to the reactor.5,7 This protocol permitted notably higher and more stable [18F]FP-CIT radiochemical yields relative to those stemming from the one-time addition of TBAOH. Nonetheless, certain limitations remained, such as the need for external base supplementation. Although we also attempted to elute trapped [18F]fluoride with a mixed elution buffer solution containing TBAOH and KOMs, fluctuating radiochemical yields were still observed as a result of variations in the amount of base added to the reactor. The current study developed two major modifications to our previous radiochemical synthesis procedure for further optimization of automated synthesis. First, we fine-tuned the procedure to allow the addition of a precise amount of the base to the trapped [18F]fluoride elution buffer solution and evaluated the radiochemical yield versus the organic impurity content as a function of the base concentration. From these results, we evaluated the optimized amount of base required for the automated synthesis of [18F]fluoromisonizadole ([18F]FMISO) with a minimal amount of precursor in the GE TRACERlab MX module. [18F]FMISO was chosen because this compound is a well-known radiopharmaceutical used to image hypoxia. [18F] FMISO is generally prepared using 5–10 mg of precursor and K2CO3 as the base, followed by purification via HPLC.8–16 Second, we introduced the ability to trap and purify [18F] FMISO via solid-phase extraction (SPE) rather than by HPLC with hydrophilic-lipophilic-balanced (HLB) polymer-based cartridges (Waters, Milford, MA, USA), a mixed-mode cation exchange (MCX) cartridge (Waters), or a Sep-Pak C18 cartridge (Waters). SPE was used because it enables the rapid and straightforward purification of radiochemicals, although SPE versus HPLC is linked to an increased possibility of high residual organic and radiochemical contamination of the final product.17 A complete analysis of organic impurities remaining after the purification of [18F]FMISO has not been undertaken before. Therefore, the present study evaluated the entire spectrum of organic impurities following [18F]FMISO synthesis and SPE purification, and compared the degree of contamination under various synthesis conditions.
Materials and Methods Materials 1-(2′-Nitro-1′-imidazolyl)-2-O-tetra-hydropyranyl-3-O-toluenesulfonyl propanediol (NITTP, the precursor for [18F]FMISO synthesis) and O-18 water (an essential reagent for [18F]fluoride production) were purchased from ABX GmbH (Dresden, Germany). All other reagents, including KOMs, were purchased from Sigma-Aldrich (St. Louis, MO, USA). HLB polymer-based, MCX ion-based, and C18 silica-based SPE cartridges were purchased from Waters. Development and optimization of manual synthesis
amount of KHCO3. Residual water was evaporated in the presence of acetonitrile (CH3CN; 500 μL) under a gentle nitrogen stream at 100°C; this process was repeated three times. The NITTP precursor (1 mg; 2.4 μmol) and CH3CN (0.5 mL) were then added to the reactor, and [18F]fluorination was carried out at 100°C for 10 min. The labeling yield of [18F]FMISO was analyzed via radio thin layer chromatography (TLC) (ethyl acetate-to-EtOH ratio = 1:1). After evaporating the CH3CN under nitrogen, hydrolysis was performed by adding HCl (1 N; 1 mL) at 100°C for 5 min. Next, neutralization was performed by adding NaOH (2 N; 0.5 mL). The crude mixture was analyzed via HPLC to identify alterations in the percentage of the organic impurities at the different base concentrations. Analytical HPLC conditions were as follows: chromatography on a Luna C18(2) column (4.6 × 250 mm, 5 μm) (Phenomenex, Torrance, CA, USA), ultraviolet (UV) detection at 320 nm, flow rate of 1 mL/min, and a mobile phase comprising H2O plus EtOH (5% EtOH for 20 min, 5%–100% EtOH for 5 min, and 100% EtOH for 5 min). Ten different organic impurities were detected on the analytical HPLC chromatogram, which were separated into two groups (groups A and B) according to the [18F]FMISO retention time. Group A impurities were characterized by retention times shorter than those of [18F]FMISO; group B impurities were characterized by retention times longer than those of [18F]FMISO. The organic impurity trends in the two groups were evaluated to determine the optimal amount of KHCO3 required for [18F]fluoride elution. Optimization of cartridge for SPE purification [18F]Fluoride (370 MBq) was trapped on a QMA cartridge and eluted with KHCO3 (33.0 μmol), 0.2 M KOMs (100 μL), kryptofix 222 (22 mg), and MeOH (600 μL). The [18F]fluorination procedure described above was used to synthesize [18F]FMISO. After synthesis, the reaction mixture was diluted with H2O (20 mL) and transferred to (1) a Sep-Pak C18 cartridge, (2) a single HLB cartridge, or (3) two serial HLB cartridges to trap [18F]FMISO. Alternatively, the reaction mixture was diluted with HCl (0.1 N; 20 mL) to protonate the [18F]FMISO, and the MCX cartridge was used as a cation exchange cartridge. Next, the amount of radioactivity deposited on the different cartridges was compared to calculate the trapping efficiency. Two serial HLB cartridges were used for the remainder of the study, as this setup afforded maximum trapping. After trapping the [18F]FMISO, 5% EtOH (2 mL; repeated five times) was applied to the serial cartridges, and each 2 mL elution was collected separately. Each of the five eluted solutions was analyzed by HPLC and examined for the presence of [18F]FMISO and organic impurities. Finally, the remaining [18F]FMISO was eluted from the cartridges using 20% EtOH (10 mL). This solution was diluted with sterile H2O (10 mL) to yield a final product with an EtOH concentration of 10%. We performed SPE purification optimization procedure with only 33.0 μmol of KHCO3 condition.
Elution buffer solution preparation The KOMs solution (0.2 M) was prepared as previously described.5 The [18F]fluoride elution buffer solution contained KOMs (0.2 M; 100 μL), MeOH (600 μL), kryptofix 222 (22 mg), and KHCO3 (9.4–33.0 μmol; 0.9–3.3 mg) as the base. Optimization of [18F]fluorination and analysis of organic impurities
732
After trapping [18F]fluoride (370 MBq) on a QMA cartridge, it was eluted with elution buffer solution containing the indicated
www.jlcr.org
Automation The GE TRACERlab MX module (GE Healthcare) was used for the automated synthesis of [18F]FMISO. Given that this module was originally designed for the production of [18F]FDG, the disposable cassette setup and synthesis sequence program were modified accordingly. Six reagent supply vials were used to carry CH3CN, the precursor, 1 N HCl, 5% EtOH, 20% EtOH, and the elution buffer solution. Furthermore, the two HLB and SPE
Copyright © 2013 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2013, 56 731–735
S. J. Lee et al. cartridges were used in tandem to trap and purify [18F]FMISO. The scheme of cassette is shown in Figure 1. After installing the disposable cassette and reagents, [18F] fluoride (3.7 GBq) was applied to the module. [18F]Fluoride was trapped on a QMA cartridge and eluted with elution buffer solution containing KHCO3 (14.1 and 33.0 μmol), KOMs (0.2 M; 100 μL), kryptofix 222 (22 mg), and MeOH (600 μL). Residual H2O was evaporated in the presence of CH3CN (150 μL); this was repeated three times. The NITTP precursor (1 mg) and CH3CN (1 mL) were then added to the reactor, and [18F] fluorination was carried out for 5 min at 100°C. After complete removal of CH3CN by evaporation, hydrolysis was performed by adding HCl (1 N; 3 mL) for 5 min at 100°C. The ensuing reaction mixture was diluted with H2O (20 mL) and transferred to the two serial HLB cartridges. Additional wash solutions containing H2O (30 mL) followed by 5% EtOH (6 mL) were used to rinse the cartridge and remove organic impurities. Purified [18F]FMISO was eluted with 20% EtOH (10 mL) and diluted with H2O (10 mL) to yield a final product with an EtOH concentration of 10%. The final product was then filter sterilized. The radiochemical yield and organic impurity content were then evaluated by HPLC, followed by gas chromatography analysis of the residual solvent. Quality control procedures (i.e., assessment of endotoxin content, radioactive half-life, and sterility, along with gamma spectroscopy to measure radionuclidic purity) were also performed to analyze the final products.
Results The [18F]fluorination yield was 34%–64% based on radio TLC analysis, with 9.4–33.0 μmol of KHCO3 used as the base (Table 1). In addition, the organic impurity contents were determined for each concentration of KHCO3 employed. The results indicated that low concentrations of KHCO3 correlated with high yields of the [18F]fluorinated product.
Table 1. [18F]FMISO yield according to base content KHCO3 (μmol) 9.4 14.1 18.9 23.6 28.3 33.0
[18F]FMISO yield (%)* 62.6 ± 4.3 63.9 ± 5.9 54.7 ± 10.9 53.5 ± 7.6 42.2 ± 6.0 34.5 ± 15.0
*Radiochemical yield from radio TLC analysis (n = 3).
High-performance liquid chromatography analysis of the crude reaction mixture revealed that the organic impurities fell into 10 different categories, IMP1 through IMP10, which were then separated into two groups (groups A and B) according to the retention time (Figure 2). Group A included polar organic impurities (IMP1 through IMP5), with a retention time shorter than that of [18F]FMISO, whereas group B included nonpolar organic impurities (IMP6 through IMP10), with a retention time longer than that of [18F]FMISO. Impurities of 80% were consisted of IMP2, IMP6, and IMP10 with both 14.1 and 23.6 μmol of KHCO3, but IMP10 and IMP6 were major impurities, respectively. However, impurities of 90% consisted of group A, and IMP1 was major impurities with 33.0 μmol of KHCO3 (Figure 3). According to the results of HPLC separation (to determine organic impurity content) and [18F]fluorination analysis (to assess radiochemical yield), we chose 14.1 and 33.0 μmol of KHCO3 as the most favorable concentrations (high radiochemical yield and suitable organic impurities pattern, respectively) for the automated synthesis of [18F]FMISO. The optimal [18F]FMISO trapping efficiency was observed with KHCO3 at a concentration of 33.0 μmol and two serial HLB cartridges. Compared with the other SPE cartridges, the serial
733
18
Figure 1. Hardware and reagent kit of GE TRACERlab MX module for [ F]FMISO production.
J. Label Compd. Radiopharm 2013, 56 731–735
Copyright © 2013 John Wiley & Sons, Ltd.
www.jlcr.org
S. J. Lee et al. 60
Area (%)
50 40 30 20 10
IMP 10
IMP 9
IMP 8
IMP 7
IMP 6
IMP 5
IMP 4
IMP 3
IMP 2
IMP 1
0
100
impurities A impurities B
Area (%)
75 50 25 0 23.6
14.1
33.0
Discussion
18
Figure 2. UV chromatogram of [ F]FMISO.
40
20 0.25
Loss of [18F]FMISO (%)
60
5.75 4.75 3.75 2.75 1.75 0.75 0.50
IMP1 IMP2 IMP4 IMP5 IMP6 [18F]FMISO
0
0.00 1
2
3
4
5
6
7
8
9
HLB cartridges showed near-complete trapping efficiency (~98%) (Table 2). The serial HLB cartridges were initially washed five times (2 mL each time) with 5% EtOH during the cartridge optimization procedure. The eluted organic impurity profiles are shown in Figure 4. IMP3 and IMP7–10 were retained in the cartridges during the SPE purification. Although very few organic impurities remained at the end of the washing procedure, a large percentage of the [18F]FMISO product was lost in the eluate. Therefore, 6 mL of 5% EtOH was used to remove organic impurities during the final automated synthesis procedure. By comparison, non-decay-corrected [18F]FMISO yields of 25.1 ± 5.0% and 13.3 ± 5.1% (n = 3) were obtained after SPE purification using automated synthesis, KHCO3 at concentrations of 14.1 and 33.0 μmol, respectively, and 1 mg of the precursor. Furthermore, the total organic impurity content after automated synthesis was 6.6 ± 2.9 ppm with 14.1 μmol KHCO3 and 1.8 ± 0.9 ppm with 33.0 μmol KHCO3. In both cases, radiochemical purity was >98%, and residual amounts of kryptofix 222 and acetonitrile were under 50.0 μg/mL and not detected, respectively. Finally, the total synthesis time was 44.3 ± 0.6 min (n = 6).
10
11
5% EtOH volume (mL) Figure 3. All kinds of impurity’s ratio (upper) and organic impurity trends in groups A and B (lower) according to base concentration.
Several reports describe the manual and automated synthesis of [18F]FMISO.8–16 In general, these reports used 5–10 mg of precursor and afforded radiochemical yields of 30%–60%. By contrast, the current study demonstrated that manual synthesis of [18F]FMISO from only 1 mg of NITTP precursor gave a [18F]incorporation yield of 64%; furthermore, automated synthesis also showed high non-decay-corrected radiochemical yields of 25% with 14 μmol KHCO3. Our results demonstrate that fine base control during [18F]fluorination, whether manual or automated, allowed for the use of reduced precursor levels, straightforward product purification, and low-cost generation of PET radiopharmaceuticals. The main advantages of KOMs are that they permit neutral pH elution of trapped [18F]fluoride and allow fine control of base content for [18F]fluorination. Bases such as K2CO3 or KHCO3 are
Table 2. Trapping efficiency of [18F]FMISO according to SPE cartridge type (n = 3) Cartridge type
Ion interaction
Hydrophobic interaction (silica-based)
SPE cartridge Trapping (%)
MCX 17.4 ± 1.2
C18 15.8 ± 5.3
Hydrophobic interaction (polymer-based) HLB 68.0 ± 9.8
HLB + HLB 98.3 ± 0.1
734
Figure 4. Organic impurity profiles according to 5% EtOH elution buffer solution volume.
www.jlcr.org
Copyright © 2013 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2013, 56 731–735
S. J. Lee et al. essential for the elution of the trapped [18F]fluoride from the ion exchange cartridge, as well as for the promotion of [18F] fluorination on the silica glass vial of the reactor. However, these bases can have negative effects on precursor compounds, including chemical decomposition and the creation of unwanted organic contaminants. HPLC purification is traditionally used to remove contaminants from radiopharmaceutical products, but this process requires long preparation times and access to an HPLC machine system fitted with an automated chemistry module. In an attempt to circumvent the need for HPLC purification, KHCO3 was added to the elution buffer solution, allowing precise control of the base concentration and the evaluation of its influence on [18F]fluorination and organic contaminant yields. Compared with high amounts of KHCO3, low amounts of base increased the [18F]fluorination yields even with minimal precursor levels. The results from the manual synthesis analysis allowed us to separate the organic impurities into two groups (Figure 3). The impurities in group A showed a shorter retention time when subjected to HPLC separation relative to [18F]FMISO, whereas the impurities in group B showed a longer retention time. Moreover, the impurities in group A had a higher polarity than both [18F]FMISO and the group B impurities and would therefore likely lend themselves to easier removal on the hydrophobic cartridges such as C18 and HLB. In addition, the group A and B impurities showed different results according to base concentration. Low amounts of base correlated with increased radiochemical yields (Table 1) and similar levels of polar group A and nonpolar B impurities (Figure 3). By contrast, high amounts of base correlated with reduced radiochemical yields; however, there was a notable reduction in the group B impurity levels, whereas the group A impurity levels were markedly increased. We anticipated that the group A impurities would also be more easily removed during the washing of HLB cartridges with 5% EtOH, whereas the group B impurities would be eluted during the product elution step with 20% EtOH. We expected to achieve high radiochemical yields with 14.1 μmol KHCO3 and high [18F]FMISO chemical purity with 33.0 μmol KHCO3. 14.1 μmol KHCO3 showed that 25% non-decay-corrected yield contained 7 ppm of organic impurities. Even though 33.0 μmol KHCO3 in automation showed a lower yield, but the amount of residual organic impurities was three times lower than 14.1 μmol. These results of automation demonstrate that our hypothesis is true. Trapping of [18F]FMISO is dependent on the characteristics of the SPE cartridges. After protonating the imidazole ring of FMISO under acidic conditions, we attempted to trap the protonated [18F]FMISO on the MCX cartridge; however, only ~17% of the product was ensnared. Compared with the MCX and the C18 cartridges (~16% product ensnarement), the HLB cartridge showed better trapping efficiency (~68% with one HLB cartridge and ~98% with two HLB cartridges in series). The present study shows two important results from automated synthesis. First, automated synthesis with 14.1 μmol KHCO3 generated a relatively high radiochemical yield, albeit with more organic impurities. Second, automated synthesis with 33.0 μmol KHCO3 reduced both the radiochemical yield and the organic impurity content. These results are interesting in light of previous reports showing that high base concentrations
generally lead to low radiochemical yields and increased organic contaminant levels. However, our results suggest that each radiopharmaceutical compound is associated with a specific optimal base concentration that produces high radiochemical yields with maximum purity and that this optimal concentration can only be determined by fine-tuning the synthesis procedures.
Conclusions The present study introduced a new base control method for producing [18F]fluoride-labeled radiopharmaceuticals. Compared with previous methods, we were able to precisely control the amount of added base using a novel elution buffer scheme incorporating KOMs. As a result, we were able to synthesize [18F]FMISO at a high yield using only 1 mg of precursor and a straightforward SPE purification step.
Acknowledgements This work was supported by a National Research Foundation of Korea grant funded by the Korean Ministry of Education, Science and Technology (grant code: 2012-0006388).
Conflict of Interest The authors did not report any conflict of interest.
References [1] J. R. Barrio, N. S. MacDonald, G. D. Robinson Jr, A. Najafi, J. S. Cook, D. E. Kuhl, J. Nucl. Med. 1981, 4, 372–375. [2] D. W. Kim, D. S. Ahn, Y. H. Oh, S. Lee, H. S. Kil, S. J. Oh, S. J. Lee, J. S. Kim, J. S. Ryu, D. H. Moon, D. Y. Chi, J. Am. Chem. Soc. 2006, 128, 16394–16397. [3] K. Hamacher, H. H. Coenen, G. Stöcklin, J. Nucl. Med. 1986, 27, 235–238. [4] M. Suehiro, S. Vallabhajosula, S. J. Goldsmith, D. J. Ballon, Appl. Radiat. Isot. 2007, 65, 1350–1358. [5] S. J. Lee, S. J. Oh, D. Y. Chi, D. H. Moon, J. S. Ryu, Bull. Korean. Chem. Soc. 2012, 33, 2177–2180. [6] J. S. Seo, B. S. Lee, S. J. Lee, S. J. Oh, D. Y. Chi, Bull. Korean. Chem. Soc. 2011, 32, 71–76. [7] S. J. Lee, S. J. Oh, W. Y. Moon, M. S. Choi, J. S. Kim, D. Y. Chi, D. H. Moon, J. S. Ryu, Nucl. Med. Biol. 2011, 38, 593–597. [8] S. J. Oh, D. Y. Chi, C. Mosdzianowski, J. Y. Kim, H. S. Gil, S. H. Kang, J. S. Ryu, D. H. Moon, Nucl. Med. Biol. 2005, 32, 899–905. [9] W. Mingwei, Z. Yingjian, Z. Yongping, Y. Huiyu, J. Radioanal. Nucl. Chem. 2009, 280, 149–155. [10] D. L. Yokell, A. K. Leece, A. Lebedev, R. Miraghaie, C. E. Ball, J. Zhang, H. Kolb, A. Elizarov, U. Mahmood, Appl. Radiat. Isot. 2012, 70, 2313–316. [11] S. Nandy, M. G. Rajan, A. Korde, N. V. Krishnamurthy, Appl. Radiat. Isot. 2010, 68, 1937–1943. [12] J. Schmaljohann, E. Schirrmacher, B. Wängler, C. Wängler, R. Schirrmacher, S. Guhlke, Nucl. Med. Biol. 2011, 38, 165–170. [13] C. W. Chang, T. K. Chou, R. S. Liu, S. J. Wang, W. J. Lin, C. H. Chen, H. E. Wang, Appl. Radiat. Isot. 2007, 65, 682–686. [14] D. Thorwarth, S. M. Eschmann, F. Paulsen, M. Alber, Phys. Med. Biol. 2005, 50, 2209–2224. [15] D. L. Yokell, A. K. Leece, L. Lebedev, R. Miraghaie, C. E. Ball, J. Zhang, H. Kolb, A. Elizarov, U. Mahmood, Appl. Radiat. Isot. 2010, 70, 2313–2316. [16] G. Tang, M. Wang, X. Tang, M. Gan, L. Luo, Nucl. Med. Biol. 2005, 32, 553–558. [17] C. Pascali, A. Bogni, L. Fugazza, C. Cucchi, O. Crispu, L. Laera, R. Iwata, G. Maiocchi, F. Crippa, E. Bombardieri, Nucl. Med. Biol. 2012, 39, 540–550.
735
J. Label Compd. Radiopharm 2013, 56 731–735
Copyright © 2013 John Wiley & Sons, Ltd.
www.jlcr.org
Revised 22 July 2013,
Accepted 18 August 2013
Published online 23 September 2013 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jlcr.3115
Development of a new precursor-minimizing base control method and its application for the automated synthesis and SPE purification of [18F]fluoromisonidazole ([18F]FMISO) Sang Ju Lee,a Ji Suk Hyun,a,b Seung Jun Oh,a* Kook Hyun Yu,b and Jin Sook Ryua Bases such as potassium carbonate and potassium bicarbonate (KHCO3) are essential for the elution of trapped [18F]fluoride from ion exchange cartridges and for the prevention of [18F]fluoride adsorption on the silica glass vial during the preparation of radiopharmaceuticals for positron emission tomography imaging. However, these bases promote the chemical decomposition of precursor compounds and the creation of unwanted organic impurities. Thus, the goal of the current study was to develop a new method for synthesizing [18F]fluoride-labeled radiopharmaceuticals (e.g., [18F] fluoromisonizadole ([18F]FMISO)) that permits the fine control of base concentrations while minimizing adverse events. Non-decay-corrected radiochemical yields of 25.1 ± 5.0% and 13.3 ± 5.1% (n = 3) were achieved after solid-phase extraction purification using automatic synthesis with GE TRACERlab MX and KHCO3 at concentrations of 14.1 and 33.0 μmol, respectively, and 1 mg of precursor (1-(2′-nitro-1′-imidazolyl)-2-O-tetra-hydropyranyl-3-O-toluenesulfonyl propanediol (NITTP)). The newly developed synthesis protocol with fine base control and low precursor content permits high radiochemical yields with minimal impurities. Keywords: [18F]fluormisonizadole; base control; [18F]fluorination; solid-phase extraction; automated synthesis; positron emission tomography
Introduction
J. Label Compd. Radiopharm 2013, 56 731–735
a Department of Nuclear Medicine, Asan Medical Center, University of Ulsan College of Medicine, 388-1 Pungnap-dong, Songpa-gu, Seoul 138-736, Korea b
Department of Chemistry, Dong Guk University, Seoul, Korea
*Correspondence to: Seung Jun Oh, Department of Nuclear Medicine, Asan Medical Center, 388-1 Pungnap-dong, Songpa-gu, Seoul 138-736, South Korea. E-mail: [email protected]
Copyright © 2013 John Wiley & Sons, Ltd.
731
Efficient [18F]fluorination of radiopharmaceuticals for positron emission tomography (PET) imaging generally requires high reaction temperatures and amount of precursor, as well as a long reaction time with [18F]fluoride, an appropriate solvent, and the catalyst. Potassium carbonate (K2CO3), potassium bicarbonate (KHCO3), and other bases are also essential for the efficient promotion of [18F]fluorination reactions because of their ability to elute trapped [18F]fluoride from ion exchange cartridges.1–3 However, the inclusion of such bases in [18F]fluorination procedures can lead to unwanted side reactions, including the decomposition of precursor, the concomitant formation of organic impurities, and necessitating a cumbersome purification step (e.g., highperformance liquid chromatography (HPLC) separation) to remove the organic contaminants. Furthermore, precursor decomposition is a common cause of low [18F]fluorinated product yields.4 The present study utilized precise base control in an attempt to overcome the complications of organic impurities contamination and low product yield. Precise control of base concentrations during [18F]fluorination procedures has a number of advantages, including the capacity to use minimal amounts of the precursor and the potential to readily purify [18F]fluorinated substrates along with a decreased occurrence of side reactions. For example, several investigations reported that control of tetrabutylammonium hydroxide (TBAOH) base levels in radiochemical synthesis reactions
reduced precursor requirements and the incidence of side reactions, while at the same time facilitating a high yield of [18F] fluorination.4–7 Nonetheless, a direct addition of TBAOH to the reactor prior to [18F]fluoride elution failed to demonstrate high radiochemical yields, the minimization of organic impurities, and straightforward application to automated synthesis systems; this was because TBAOH added in this manner could not effectively elute [18F]fluoride from the ion exchange cartridge.4 We previously developed a novel protocol to elute [18F]fluoride with a neutral salt solution and applied this methodology to the automated synthesis of [18F]fluoropropyl-carbomethoxyiodophenyl-nor-tropane ([18F]FP-CIT), a radioactive tracer.7 We hypothesized that a salt-based elution of trapped [18F]fluoride would allow for the additional direct application of base to the reactor and the fine control of base concentrations. According to this hypothesis, [18F]fluoride was eluted with a potassium methanesulfonate (KOMs) solution, and a precise amount of
S. J. Lee et al. TBAOH was then added to the reactor.5,7 This protocol permitted notably higher and more stable [18F]FP-CIT radiochemical yields relative to those stemming from the one-time addition of TBAOH. Nonetheless, certain limitations remained, such as the need for external base supplementation. Although we also attempted to elute trapped [18F]fluoride with a mixed elution buffer solution containing TBAOH and KOMs, fluctuating radiochemical yields were still observed as a result of variations in the amount of base added to the reactor. The current study developed two major modifications to our previous radiochemical synthesis procedure for further optimization of automated synthesis. First, we fine-tuned the procedure to allow the addition of a precise amount of the base to the trapped [18F]fluoride elution buffer solution and evaluated the radiochemical yield versus the organic impurity content as a function of the base concentration. From these results, we evaluated the optimized amount of base required for the automated synthesis of [18F]fluoromisonizadole ([18F]FMISO) with a minimal amount of precursor in the GE TRACERlab MX module. [18F]FMISO was chosen because this compound is a well-known radiopharmaceutical used to image hypoxia. [18F] FMISO is generally prepared using 5–10 mg of precursor and K2CO3 as the base, followed by purification via HPLC.8–16 Second, we introduced the ability to trap and purify [18F] FMISO via solid-phase extraction (SPE) rather than by HPLC with hydrophilic-lipophilic-balanced (HLB) polymer-based cartridges (Waters, Milford, MA, USA), a mixed-mode cation exchange (MCX) cartridge (Waters), or a Sep-Pak C18 cartridge (Waters). SPE was used because it enables the rapid and straightforward purification of radiochemicals, although SPE versus HPLC is linked to an increased possibility of high residual organic and radiochemical contamination of the final product.17 A complete analysis of organic impurities remaining after the purification of [18F]FMISO has not been undertaken before. Therefore, the present study evaluated the entire spectrum of organic impurities following [18F]FMISO synthesis and SPE purification, and compared the degree of contamination under various synthesis conditions.
Materials and Methods Materials 1-(2′-Nitro-1′-imidazolyl)-2-O-tetra-hydropyranyl-3-O-toluenesulfonyl propanediol (NITTP, the precursor for [18F]FMISO synthesis) and O-18 water (an essential reagent for [18F]fluoride production) were purchased from ABX GmbH (Dresden, Germany). All other reagents, including KOMs, were purchased from Sigma-Aldrich (St. Louis, MO, USA). HLB polymer-based, MCX ion-based, and C18 silica-based SPE cartridges were purchased from Waters. Development and optimization of manual synthesis
amount of KHCO3. Residual water was evaporated in the presence of acetonitrile (CH3CN; 500 μL) under a gentle nitrogen stream at 100°C; this process was repeated three times. The NITTP precursor (1 mg; 2.4 μmol) and CH3CN (0.5 mL) were then added to the reactor, and [18F]fluorination was carried out at 100°C for 10 min. The labeling yield of [18F]FMISO was analyzed via radio thin layer chromatography (TLC) (ethyl acetate-to-EtOH ratio = 1:1). After evaporating the CH3CN under nitrogen, hydrolysis was performed by adding HCl (1 N; 1 mL) at 100°C for 5 min. Next, neutralization was performed by adding NaOH (2 N; 0.5 mL). The crude mixture was analyzed via HPLC to identify alterations in the percentage of the organic impurities at the different base concentrations. Analytical HPLC conditions were as follows: chromatography on a Luna C18(2) column (4.6 × 250 mm, 5 μm) (Phenomenex, Torrance, CA, USA), ultraviolet (UV) detection at 320 nm, flow rate of 1 mL/min, and a mobile phase comprising H2O plus EtOH (5% EtOH for 20 min, 5%–100% EtOH for 5 min, and 100% EtOH for 5 min). Ten different organic impurities were detected on the analytical HPLC chromatogram, which were separated into two groups (groups A and B) according to the [18F]FMISO retention time. Group A impurities were characterized by retention times shorter than those of [18F]FMISO; group B impurities were characterized by retention times longer than those of [18F]FMISO. The organic impurity trends in the two groups were evaluated to determine the optimal amount of KHCO3 required for [18F]fluoride elution. Optimization of cartridge for SPE purification [18F]Fluoride (370 MBq) was trapped on a QMA cartridge and eluted with KHCO3 (33.0 μmol), 0.2 M KOMs (100 μL), kryptofix 222 (22 mg), and MeOH (600 μL). The [18F]fluorination procedure described above was used to synthesize [18F]FMISO. After synthesis, the reaction mixture was diluted with H2O (20 mL) and transferred to (1) a Sep-Pak C18 cartridge, (2) a single HLB cartridge, or (3) two serial HLB cartridges to trap [18F]FMISO. Alternatively, the reaction mixture was diluted with HCl (0.1 N; 20 mL) to protonate the [18F]FMISO, and the MCX cartridge was used as a cation exchange cartridge. Next, the amount of radioactivity deposited on the different cartridges was compared to calculate the trapping efficiency. Two serial HLB cartridges were used for the remainder of the study, as this setup afforded maximum trapping. After trapping the [18F]FMISO, 5% EtOH (2 mL; repeated five times) was applied to the serial cartridges, and each 2 mL elution was collected separately. Each of the five eluted solutions was analyzed by HPLC and examined for the presence of [18F]FMISO and organic impurities. Finally, the remaining [18F]FMISO was eluted from the cartridges using 20% EtOH (10 mL). This solution was diluted with sterile H2O (10 mL) to yield a final product with an EtOH concentration of 10%. We performed SPE purification optimization procedure with only 33.0 μmol of KHCO3 condition.
Elution buffer solution preparation The KOMs solution (0.2 M) was prepared as previously described.5 The [18F]fluoride elution buffer solution contained KOMs (0.2 M; 100 μL), MeOH (600 μL), kryptofix 222 (22 mg), and KHCO3 (9.4–33.0 μmol; 0.9–3.3 mg) as the base. Optimization of [18F]fluorination and analysis of organic impurities
732
After trapping [18F]fluoride (370 MBq) on a QMA cartridge, it was eluted with elution buffer solution containing the indicated
www.jlcr.org
Automation The GE TRACERlab MX module (GE Healthcare) was used for the automated synthesis of [18F]FMISO. Given that this module was originally designed for the production of [18F]FDG, the disposable cassette setup and synthesis sequence program were modified accordingly. Six reagent supply vials were used to carry CH3CN, the precursor, 1 N HCl, 5% EtOH, 20% EtOH, and the elution buffer solution. Furthermore, the two HLB and SPE
Copyright © 2013 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2013, 56 731–735
S. J. Lee et al. cartridges were used in tandem to trap and purify [18F]FMISO. The scheme of cassette is shown in Figure 1. After installing the disposable cassette and reagents, [18F] fluoride (3.7 GBq) was applied to the module. [18F]Fluoride was trapped on a QMA cartridge and eluted with elution buffer solution containing KHCO3 (14.1 and 33.0 μmol), KOMs (0.2 M; 100 μL), kryptofix 222 (22 mg), and MeOH (600 μL). Residual H2O was evaporated in the presence of CH3CN (150 μL); this was repeated three times. The NITTP precursor (1 mg) and CH3CN (1 mL) were then added to the reactor, and [18F] fluorination was carried out for 5 min at 100°C. After complete removal of CH3CN by evaporation, hydrolysis was performed by adding HCl (1 N; 3 mL) for 5 min at 100°C. The ensuing reaction mixture was diluted with H2O (20 mL) and transferred to the two serial HLB cartridges. Additional wash solutions containing H2O (30 mL) followed by 5% EtOH (6 mL) were used to rinse the cartridge and remove organic impurities. Purified [18F]FMISO was eluted with 20% EtOH (10 mL) and diluted with H2O (10 mL) to yield a final product with an EtOH concentration of 10%. The final product was then filter sterilized. The radiochemical yield and organic impurity content were then evaluated by HPLC, followed by gas chromatography analysis of the residual solvent. Quality control procedures (i.e., assessment of endotoxin content, radioactive half-life, and sterility, along with gamma spectroscopy to measure radionuclidic purity) were also performed to analyze the final products.
Results The [18F]fluorination yield was 34%–64% based on radio TLC analysis, with 9.4–33.0 μmol of KHCO3 used as the base (Table 1). In addition, the organic impurity contents were determined for each concentration of KHCO3 employed. The results indicated that low concentrations of KHCO3 correlated with high yields of the [18F]fluorinated product.
Table 1. [18F]FMISO yield according to base content KHCO3 (μmol) 9.4 14.1 18.9 23.6 28.3 33.0
[18F]FMISO yield (%)* 62.6 ± 4.3 63.9 ± 5.9 54.7 ± 10.9 53.5 ± 7.6 42.2 ± 6.0 34.5 ± 15.0
*Radiochemical yield from radio TLC analysis (n = 3).
High-performance liquid chromatography analysis of the crude reaction mixture revealed that the organic impurities fell into 10 different categories, IMP1 through IMP10, which were then separated into two groups (groups A and B) according to the retention time (Figure 2). Group A included polar organic impurities (IMP1 through IMP5), with a retention time shorter than that of [18F]FMISO, whereas group B included nonpolar organic impurities (IMP6 through IMP10), with a retention time longer than that of [18F]FMISO. Impurities of 80% were consisted of IMP2, IMP6, and IMP10 with both 14.1 and 23.6 μmol of KHCO3, but IMP10 and IMP6 were major impurities, respectively. However, impurities of 90% consisted of group A, and IMP1 was major impurities with 33.0 μmol of KHCO3 (Figure 3). According to the results of HPLC separation (to determine organic impurity content) and [18F]fluorination analysis (to assess radiochemical yield), we chose 14.1 and 33.0 μmol of KHCO3 as the most favorable concentrations (high radiochemical yield and suitable organic impurities pattern, respectively) for the automated synthesis of [18F]FMISO. The optimal [18F]FMISO trapping efficiency was observed with KHCO3 at a concentration of 33.0 μmol and two serial HLB cartridges. Compared with the other SPE cartridges, the serial
733
18
Figure 1. Hardware and reagent kit of GE TRACERlab MX module for [ F]FMISO production.
J. Label Compd. Radiopharm 2013, 56 731–735
Copyright © 2013 John Wiley & Sons, Ltd.
www.jlcr.org
S. J. Lee et al. 60
Area (%)
50 40 30 20 10
IMP 10
IMP 9
IMP 8
IMP 7
IMP 6
IMP 5
IMP 4
IMP 3
IMP 2
IMP 1
0
100
impurities A impurities B
Area (%)
75 50 25 0 23.6
14.1
33.0
Discussion
18
Figure 2. UV chromatogram of [ F]FMISO.
40
20 0.25
Loss of [18F]FMISO (%)
60
5.75 4.75 3.75 2.75 1.75 0.75 0.50
IMP1 IMP2 IMP4 IMP5 IMP6 [18F]FMISO
0
0.00 1
2
3
4
5
6
7
8
9
HLB cartridges showed near-complete trapping efficiency (~98%) (Table 2). The serial HLB cartridges were initially washed five times (2 mL each time) with 5% EtOH during the cartridge optimization procedure. The eluted organic impurity profiles are shown in Figure 4. IMP3 and IMP7–10 were retained in the cartridges during the SPE purification. Although very few organic impurities remained at the end of the washing procedure, a large percentage of the [18F]FMISO product was lost in the eluate. Therefore, 6 mL of 5% EtOH was used to remove organic impurities during the final automated synthesis procedure. By comparison, non-decay-corrected [18F]FMISO yields of 25.1 ± 5.0% and 13.3 ± 5.1% (n = 3) were obtained after SPE purification using automated synthesis, KHCO3 at concentrations of 14.1 and 33.0 μmol, respectively, and 1 mg of the precursor. Furthermore, the total organic impurity content after automated synthesis was 6.6 ± 2.9 ppm with 14.1 μmol KHCO3 and 1.8 ± 0.9 ppm with 33.0 μmol KHCO3. In both cases, radiochemical purity was >98%, and residual amounts of kryptofix 222 and acetonitrile were under 50.0 μg/mL and not detected, respectively. Finally, the total synthesis time was 44.3 ± 0.6 min (n = 6).
10
11
5% EtOH volume (mL) Figure 3. All kinds of impurity’s ratio (upper) and organic impurity trends in groups A and B (lower) according to base concentration.
Several reports describe the manual and automated synthesis of [18F]FMISO.8–16 In general, these reports used 5–10 mg of precursor and afforded radiochemical yields of 30%–60%. By contrast, the current study demonstrated that manual synthesis of [18F]FMISO from only 1 mg of NITTP precursor gave a [18F]incorporation yield of 64%; furthermore, automated synthesis also showed high non-decay-corrected radiochemical yields of 25% with 14 μmol KHCO3. Our results demonstrate that fine base control during [18F]fluorination, whether manual or automated, allowed for the use of reduced precursor levels, straightforward product purification, and low-cost generation of PET radiopharmaceuticals. The main advantages of KOMs are that they permit neutral pH elution of trapped [18F]fluoride and allow fine control of base content for [18F]fluorination. Bases such as K2CO3 or KHCO3 are
Table 2. Trapping efficiency of [18F]FMISO according to SPE cartridge type (n = 3) Cartridge type
Ion interaction
Hydrophobic interaction (silica-based)
SPE cartridge Trapping (%)
MCX 17.4 ± 1.2
C18 15.8 ± 5.3
Hydrophobic interaction (polymer-based) HLB 68.0 ± 9.8
HLB + HLB 98.3 ± 0.1
734
Figure 4. Organic impurity profiles according to 5% EtOH elution buffer solution volume.
www.jlcr.org
Copyright © 2013 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2013, 56 731–735
S. J. Lee et al. essential for the elution of the trapped [18F]fluoride from the ion exchange cartridge, as well as for the promotion of [18F] fluorination on the silica glass vial of the reactor. However, these bases can have negative effects on precursor compounds, including chemical decomposition and the creation of unwanted organic contaminants. HPLC purification is traditionally used to remove contaminants from radiopharmaceutical products, but this process requires long preparation times and access to an HPLC machine system fitted with an automated chemistry module. In an attempt to circumvent the need for HPLC purification, KHCO3 was added to the elution buffer solution, allowing precise control of the base concentration and the evaluation of its influence on [18F]fluorination and organic contaminant yields. Compared with high amounts of KHCO3, low amounts of base increased the [18F]fluorination yields even with minimal precursor levels. The results from the manual synthesis analysis allowed us to separate the organic impurities into two groups (Figure 3). The impurities in group A showed a shorter retention time when subjected to HPLC separation relative to [18F]FMISO, whereas the impurities in group B showed a longer retention time. Moreover, the impurities in group A had a higher polarity than both [18F]FMISO and the group B impurities and would therefore likely lend themselves to easier removal on the hydrophobic cartridges such as C18 and HLB. In addition, the group A and B impurities showed different results according to base concentration. Low amounts of base correlated with increased radiochemical yields (Table 1) and similar levels of polar group A and nonpolar B impurities (Figure 3). By contrast, high amounts of base correlated with reduced radiochemical yields; however, there was a notable reduction in the group B impurity levels, whereas the group A impurity levels were markedly increased. We anticipated that the group A impurities would also be more easily removed during the washing of HLB cartridges with 5% EtOH, whereas the group B impurities would be eluted during the product elution step with 20% EtOH. We expected to achieve high radiochemical yields with 14.1 μmol KHCO3 and high [18F]FMISO chemical purity with 33.0 μmol KHCO3. 14.1 μmol KHCO3 showed that 25% non-decay-corrected yield contained 7 ppm of organic impurities. Even though 33.0 μmol KHCO3 in automation showed a lower yield, but the amount of residual organic impurities was three times lower than 14.1 μmol. These results of automation demonstrate that our hypothesis is true. Trapping of [18F]FMISO is dependent on the characteristics of the SPE cartridges. After protonating the imidazole ring of FMISO under acidic conditions, we attempted to trap the protonated [18F]FMISO on the MCX cartridge; however, only ~17% of the product was ensnared. Compared with the MCX and the C18 cartridges (~16% product ensnarement), the HLB cartridge showed better trapping efficiency (~68% with one HLB cartridge and ~98% with two HLB cartridges in series). The present study shows two important results from automated synthesis. First, automated synthesis with 14.1 μmol KHCO3 generated a relatively high radiochemical yield, albeit with more organic impurities. Second, automated synthesis with 33.0 μmol KHCO3 reduced both the radiochemical yield and the organic impurity content. These results are interesting in light of previous reports showing that high base concentrations
generally lead to low radiochemical yields and increased organic contaminant levels. However, our results suggest that each radiopharmaceutical compound is associated with a specific optimal base concentration that produces high radiochemical yields with maximum purity and that this optimal concentration can only be determined by fine-tuning the synthesis procedures.
Conclusions The present study introduced a new base control method for producing [18F]fluoride-labeled radiopharmaceuticals. Compared with previous methods, we were able to precisely control the amount of added base using a novel elution buffer scheme incorporating KOMs. As a result, we were able to synthesize [18F]FMISO at a high yield using only 1 mg of precursor and a straightforward SPE purification step.
Acknowledgements This work was supported by a National Research Foundation of Korea grant funded by the Korean Ministry of Education, Science and Technology (grant code: 2012-0006388).
Conflict of Interest The authors did not report any conflict of interest.
References [1] J. R. Barrio, N. S. MacDonald, G. D. Robinson Jr, A. Najafi, J. S. Cook, D. E. Kuhl, J. Nucl. Med. 1981, 4, 372–375. [2] D. W. Kim, D. S. Ahn, Y. H. Oh, S. Lee, H. S. Kil, S. J. Oh, S. J. Lee, J. S. Kim, J. S. Ryu, D. H. Moon, D. Y. Chi, J. Am. Chem. Soc. 2006, 128, 16394–16397. [3] K. Hamacher, H. H. Coenen, G. Stöcklin, J. Nucl. Med. 1986, 27, 235–238. [4] M. Suehiro, S. Vallabhajosula, S. J. Goldsmith, D. J. Ballon, Appl. Radiat. Isot. 2007, 65, 1350–1358. [5] S. J. Lee, S. J. Oh, D. Y. Chi, D. H. Moon, J. S. Ryu, Bull. Korean. Chem. Soc. 2012, 33, 2177–2180. [6] J. S. Seo, B. S. Lee, S. J. Lee, S. J. Oh, D. Y. Chi, Bull. Korean. Chem. Soc. 2011, 32, 71–76. [7] S. J. Lee, S. J. Oh, W. Y. Moon, M. S. Choi, J. S. Kim, D. Y. Chi, D. H. Moon, J. S. Ryu, Nucl. Med. Biol. 2011, 38, 593–597. [8] S. J. Oh, D. Y. Chi, C. Mosdzianowski, J. Y. Kim, H. S. Gil, S. H. Kang, J. S. Ryu, D. H. Moon, Nucl. Med. Biol. 2005, 32, 899–905. [9] W. Mingwei, Z. Yingjian, Z. Yongping, Y. Huiyu, J. Radioanal. Nucl. Chem. 2009, 280, 149–155. [10] D. L. Yokell, A. K. Leece, A. Lebedev, R. Miraghaie, C. E. Ball, J. Zhang, H. Kolb, A. Elizarov, U. Mahmood, Appl. Radiat. Isot. 2012, 70, 2313–316. [11] S. Nandy, M. G. Rajan, A. Korde, N. V. Krishnamurthy, Appl. Radiat. Isot. 2010, 68, 1937–1943. [12] J. Schmaljohann, E. Schirrmacher, B. Wängler, C. Wängler, R. Schirrmacher, S. Guhlke, Nucl. Med. Biol. 2011, 38, 165–170. [13] C. W. Chang, T. K. Chou, R. S. Liu, S. J. Wang, W. J. Lin, C. H. Chen, H. E. Wang, Appl. Radiat. Isot. 2007, 65, 682–686. [14] D. Thorwarth, S. M. Eschmann, F. Paulsen, M. Alber, Phys. Med. Biol. 2005, 50, 2209–2224. [15] D. L. Yokell, A. K. Leece, L. Lebedev, R. Miraghaie, C. E. Ball, J. Zhang, H. Kolb, A. Elizarov, U. Mahmood, Appl. Radiat. Isot. 2010, 70, 2313–2316. [16] G. Tang, M. Wang, X. Tang, M. Gan, L. Luo, Nucl. Med. Biol. 2005, 32, 553–558. [17] C. Pascali, A. Bogni, L. Fugazza, C. Cucchi, O. Crispu, L. Laera, R. Iwata, G. Maiocchi, F. Crippa, E. Bombardieri, Nucl. Med. Biol. 2012, 39, 540–550.
735
J. Label Compd. Radiopharm 2013, 56 731–735
Copyright © 2013 John Wiley & Sons, Ltd.
www.jlcr.org
![Fully automated synthesis of [(18) F]fluoro-dihydrotestosterone ([(18) F]FDHT) using the FlexLab module.](/assets/img/hold.png)
