Reagents
No-carrier-added [18F]fluoride was produced by the 18O(p, n)18F reaction from [18O]H2O (84% isotopic purity, Zevacor Pharma, Noblesville, IN, USA) in an RDS-112 cyclotron (Siemens; Knoxville, TN, USA) at 11 MeV using a 1 mL tantalum target with havar foil. Acetonitrile (MeCN; anhydrous, 99.8%), methanol (MeOH; anhydrous, 99.8%), ethanol (EtOH; 200 proof, >99.5%), hydrochloric acid (HCl; 1M), dimethylsulfoxide (DMSO, 98%), deionized (DI) water, and polyethylene glycol 400 (PEG 400) were purchased from Millipore Sigma (St. Louis, MO, USA). All reagents were used as received without further purification. N-Boc protected mesylate FBB precursor ([methanesulfonic acid 2-{2-[2-(4-{2-[4-(tert-butoxycarbonyl-methyl-amino)-phenyl]-vinyl}-phenoxy)-ethoxy]-ethoxy}-ethyl ester) and FBB reference standard (4-[(E)-2-(4-{2-[2-(2-[18F]fluoroethoxy) ethoxy] ethoxy} phenyl) vinyl]-N-methylaniline) were generously provided by Life Molecular Imaging GmbH as a part of [18F]Florbetaben synthesis kits. Kryptofix 222 (K222) and potassium carbonate (K2CO3) were purchased from ABX GmbH (Radeberg, Germany). Sodium phosphate dibasic (Na2HPO4-7H2O) and sodium phosphate monobasic (NaH2PO4H2O) were purchased from Fisher Scientific (Thermo Fisher Scientific, Waltham, MA). Ultrapure 18 MΩ water was acquired through a Milli-Q Integral 3 purification system (Millipore Sigma, St. Louis, MO, USA).
Scavenger mix (used in multiple steps of the reaction), consisting of sodium ascorbate with L-ascorbic acid (87:13 w/w), was obtained from the [18F]Florbetaben production kits provided by Life Molecular Imaging GmbH. HPLC mobile phase was prepared by first dissolving 1.785 g of Na2HPO4-7H2O and 0.461 g of NaH2PO4H2O in 0.40 L of 18 MΩ H2O to make 25 mM phosphate buffer (pH 7.2), then adding in 0.60 L of MeCN. Collection mixture to recover the crude [18F]Florbetaben from the chip consisted of MeCN mixed with 33 mg/mL scavenger mix in DI water (1:1, v/v). Stabilization / dilution solution used in formulation consisted of 200 µL PEG 400 and 650 µL DI water with 33 mg of scavenger mix.
Analytical methods
A calibrated ion chamber (CRC 25-PET, Capintec, Florham Park, NJ, USA) was used to perform radioactivity measurements. For radio-thin-layer chromatography (radio-TLC) analysis, reverse phase TLC plates (RP-18 silica gel 60 F254 sheets; aluminum backing; Millipore Sigma, St. Louis, MO, USA) were cut into 15 x 60 mm pieces (with 40 mm developing distance), spotted with 1 µL of the sample and developed in 90% (v/v) MeCN in H2O. TLC plates were analyzed with a Cerenkov luminescence imaging system as previously described [23] or a conventional radio-TLC scanner (miniGita star, Raytest, Inc., Wilmington, NC, USA). Retention factors of the observed radioactive species were: 0.0 ([18F]fluoride), 0.4 ([18F]FBB), and 0.8 (fluorinated intermediate).
Radio-HPLC analysis and purification were performed on an analytical-scale Smartline HPLC system (Knauer, Berlin, Germany) with 200 µL injection loop, a pump (Model 1000), degasser (Model 5050), UV detector (Model 2500) and a radiometric detector (Bioscan B-FC-4000, Bioscan Inc., Washington DC, USA). Samples were separated using a C18 column (Luna, 5 µm particles, 100Å pores, 250 x 4.6 mm, Phenomenex, Torrance, CA, USA) with guard column (SecurityGuard C18, Phenomenex). UV absorbance was measured at 254 nm. Using isocratic conditions with a MeCN : 25 mM phosphate buffer 60:40 (v/v) mobile phase delivered at 1.5 mL/min, the observed retention time of [18F]fluoride was between 2-3 min, 6 min for [18F]FBB, and 14 min for the fluorinated intermediate.
Multiple measurements were collected during the synthesis to calculate several parameters. Unless otherwise specified, all percentage values (yields, efficiencies) are decay-corrected (d.c.). Starting activity was determined by calculating a difference in activity measurements of a source vial before and after addition of the radionuclide from the vial to the chip (accounting for losses in pipette tips). Collection efficiency is the ratio of activity of the crude reaction mixture recovered from the chip relative to the starting activity. Residual chip activity is the percentage of starting activity that remained on the chip after the synthesis and crude reaction mixture recovery. Radiochemical conversion is the percentage of the desired product ([18F]FBB) in the crude mixture as determined by radio-TLC. Crude [18F]FBB radiochemical yield (crude RCY) is calculated by multiplying the collection efficiency by the radiochemical conversion. Isolated yield is the ratio of the activity of the purified product collected after HPLC purification to the starting activity. Formulated product yield is calculated by dividing the activity of the final formulated product by the starting activity.
To carefully evaluate the formulation performance, additional parameters were calculated relative to the activity of pure [18F]FBB fraction obtained after HPLC purification. The formulation efficiency is the ratio of formulated product activity to the activity of the [18F]FBB fraction. Activity in waste is the ratio of activity in the waste container relative to the [18F]FBB fraction, and fraction collection vial residual activity is the percentage of activity remaining in the initial [18F]FBB fraction vial after the formulation process is complete. Cartridge residual activity is the percentage of the initial [18F]FBB activity that remained on the cartridge after formulation. Residual in the system is the percentage of the initial [18F]FBB fraction that was not recovered (i.e. remaining in various portions of the formulation system, e.g. valves, tubing, etc.).
Droplet synthesis platform
Radiochemistry was performed in droplet format using Teflon-coated silicon chips that had small circular regions of Teflon etched away, leaving hydrophilic patches that act as surface-tension traps to confine reagents during the multi-step radiosynthesis. Temperature control was achieved by affixing the chip atop a ceramic heater with thermal paste. The details of the chip fabrication were previously reported [10]. Initially, the conditions were optimized using chips containing 4 reaction sites [24] on a platform with 4 heaters. Based on optimized conditions, the synthesis was adapted onto an ultra-compact automated droplet radiosynthesizer [13] allowing for reduced radiation exposure and operation time.
The overall setup comprises a droplet synthesizer, analytical-scale HPLC purification, and a newly-developed automated solid-phase extraction setup to perform formulation using custom micro-cartridges with C18 resin (Figure 1). Details of the formulation system and cartridge fabrication are provided in the Supporting information, sections 1 and 2, respectively. Briefly, the inlet of the cartridge was connected to a selector valve and the outlet to a 3-way valve. Using the selector valve, different solutions could be flowed through the cartridge such as the [18F]FBB fraction vial (trapping step), a vial with aqueous sodium ascorbate solution (washing step), and a vial with ethanol (elution step). The 3-way valve was used to direct the cartridge output to waste (trapping and washing steps) or the product vial (elution step). The liquid movement was initiated by applying nitrogen pressure to the vials containing [18F]FBB (15 psi), water (15 psi) and ethanol (3 psi). A program written in LabView (National Instruments, Austin, TX) automatically controlled the valves and pressure sources via a data acquisition module (DAQ) to complete the trapping, washing, and elution steps.
Microvolume radiosynthesis
Preliminary synthesis optimization
The microvolume synthesis was adapted from the common 2-step approach, which consists of fluorination of the Boc-protected precursor using [18F]KF/K222, followed by a hydrolysis step [5]. A schematic representation of the microvolume synthesis process is shown in Figure 2.
Initial pre-optimization conditions for microvolume synthesis were selected by scaling down the conventional synthesis conditions reported in Collins et al. [25]. The fluorination reaction volume was reduced 90-fold from 1.8 mL to 20 µL, but the precursor concentration was maintained at 4 mM, resulting in a precursor amount of 120 nmol. The fluorination solvent was changed from MeCN to DMSO, since MeCN evaporated too quickly in droplet format. The total amount of cryptand phase-transfer catalyst was reduced 180-fold (from 49 µmol to 275 nmol of K2CO3 and from 68 µmol to 383 nmol of K222). First, the effect of temperature on the fluorination reaction was studied, followed by optimization of the amount of K2CO3/K222, and then amount of precursor. Each set of conditions was repeated n = 4 times, with reagents delivered manually via pipette.
For each experiment, aqueous [18F]fluoride (10-20 µL; ~7.4 - 370 MBq [~0.2 - 10 mCi]) was mixed with the desired amount of K222/K2CO3 in 4.5 µL H2O and loaded to the reaction site to be evaporated to dryness at 100 °C for 2 min. Next the desired amount of precursor in DMSO was added to the dried fluoride residue and reacted at the desired temperature for 5 min. For initial optimization experiments the crude product of the fluorination reaction was collected and analyzed. In other cases, the hydrolysis step was performed by adding 20 μL of 1N HCl to the reaction mixture and heating at 90 °C for 3 mins. To recover the crude product (or intermediate), 20 µL of collection mixture was added to resuspend the product on chip, and then transferred into the crude product vial. To ensure thorough recovery from the chip, the collection procedure was repeated 2 more times (3 more times for the automated setup). To avoid radiolysis and photodegradation, the crude product vial was preloaded with 64 µL of water with 33 mg/mL of scavenger mix and kept in the dark.
Automated microvolume synthesis
The automated synthesis of [18F]FBB was performed using identical chips, but using a custom-built platform [13] that supported automated reagent dispensing and product recovery. The reaction conditions were identical to the optimized manual synthesis conditions, except that the deprotection was performed using 1M HCl:MeCN 1:1 (v/v). This 20 µL acidic mixture was dispensed at the beginning of deprotection and another 20 μL after 1.5 min. The diluted acid was used to reduce damage to the reagent dispensers.
Purification and formulation
To perform purification, the crude product collected in aqueous scavenger solution was diluted with aqueous sodium phosphate buffer to a total volume of 175 µL and delivered into an analytical radio-HPLC system with 200 µL injection loop. The [18F]FBB peak (retention time 6 min) was collected (for 1.0 – 1.5 min) into a 50 mL conical tube (Falcon, Corning, USA) pre-loaded with 33 mg/mL scavenger mix in 3 mL water and covered by aluminum foil.
Formulation was performed by diluting the purified [18F]FBB with 30 mL of DI water, and carrying out solid phase extraction (SPE) using a C18 cartridge. Initially, commercial C18 cartridges (Waters Sep-pak C18 Plus Light, 130 mg, Waters Corporation, Milford, MA) were used, but eventually the use of custom miniature cartridges made by packing C18 resin into lengths of tubing (Supporting Information, Section 1) was explored to allow reduction of the final formulated volume. Cartridges made inside 0.02” ID tubing exhibited extremely low flow rates, but cartridges packed inside 0.0625” ID tubing had suitable flow rates. Preconditioning of miniature cartridges was performed with 5 mL MeOH followed by 6 mL of DI water at approximately 1 mL/min. Preliminary testing of the cartridges was performed manually using disposable syringes, and later the procedure was automated using the automated solid-phase extraction setup (Figure 1).
In the final formulation procedure, the diluted [18F]FBB was trapped on the cartridge, the cartridge was then washed by flowing through 10 mL DI water containing 10 mg/mL of scavenger mix to remove residual solvents and impurities (The amount of scavenger is the same as reported by Rominger et al. [20]). Finally, the trapped [18F]FBB was eluted from the cartridge using 150 µL EtOH into an amber-colored glass product vial preloaded with 850 µL of stabilization solution.