Abstract
l-p-Boronophenylalanine (BPA) has been applied as a potential boron carrier for the treatment of malignant glioma in clinical boron neutron capture therapy (BNCT) since 1994. To provide the pharmacokinetics of BPA for clinical use of BNCT in Taiwan, 4-borono-2-18F-fluoro-L-phenylalanine-fructose (18F-FBPA-Fr) was synthesized and the biologic characteristics of this radiotracer in glioma-bearing rats were investigated. Methods: Radiolabeled 18F-F2 was produced via the 20Ne(d,α)18F reaction, and 18F-acetyl hypofluorite (18F-AcOF) was generated by passing 18F-F2 through a column filled with tightly packed KOAc/HOAc powder. The effluent containing 18F-AcOF was bubbled into BPA in trifluoroacetic acid, then purified by high-performance liquid chromatography, and further composited with fructose to afford 18F-FBPA-Fr. Male Fischer 344 rats bearing F98 glioma in the left brain were used for biologic studies. The biodistribution of BPA-Fr and 18F-FBPA-Fr was determined, and the microautoradiography and PET imaging of 18F-FBPA-Fr were performed, on the 13th day after tumor inoculation. Results: The radiochemical purity of 18F-FBPA-Fr was >97% and the radiochemical yield of 18F-FBPA-Fr was 20%–25%. In glioma-bearing rats, the accumulation ratios of B-10 for glioma-to-normal brain were 2.05, 1.86, 1.24, and 1.10 at 0.5, 1, 2, and 4 h, respectively, after administration of 43 mg BPA-Fr via the tail vein. The accumulation ratios of 18F-FBPA-Fr for glioma-to-normal brain were 3.45, 3.13, 2.61, and 2.02, whereas the tumor-to-heart blood ratios were 1.72, 2.61, 2.00, and 1.93, respectively, for the same time points. The uptake characteristics of BPA-Fr and 18F-FBPA-Fr in F98 glioma were similar with a maximum at 1 h after the drugs’ administration. The results obtained from the biodistribution studies indicated that 0.5–1 h after BPA-Fr injection would be the optimal time for BNCT. Biodistribution, PET images, and brain microautoradiography of 18F-FBPA-Fr all confirmed this finding. Conclusion: 18F-FBPA-Fr showed specific tumor uptake in F98 glioma-bearing rats and could be used as a probe for BPA-Fr in BNCT. This study provides useful information for the future clinical application of BNCT in brain tumor therapy.
- boron neutron capture therapy
- 4-borono-2-18F-fluoro-l-phenylalanine-fructose
- F98 glioma
- microautoradiography
- PET
The basic concept of a binary system for cancer treatment is to selectively destroy malignant cells while concomitantly sparing normal tissues (1,2). Boron neutron capture therapy (BNCT) is one of the binary cancer treatment systems that is based on the selective accumulation of 10B in tumors and then irradiation with a neutron source. The selective accumulation of 10B in tumors and the subsequent capture of an epithermal neutron by a 10B atom eject an α-particle (He2+) and a lithium nucleus (7Li) in opposite directions. The average track of these densely ionizing particles is approximately 14 μm, about the diameter of one cell, so that killing of tumor cells is highly efficient (3). BNCT has been proposed for the treatment of human gliomas since 1936 (4). Clinical trials are presently conducted in the United States, Japan, and Europe using reactor-generated epithermal neutron beams and 10B-containing compounds such as sulfhydryl borane (Na2B12H11SH) and l-p-boronophenylalanine (BPA) (5–7).
In Taiwan, the reactor at National Tsing-Hua University (THOR) is under remodeling and will be upgraded as a dedicated facility for BNCT with an ultimate goal to perform clinical trials for the treatment of gliomas and hepatomas. The success of BNCT depends mainly on the differential uptake of the boronated compound in tumors compared with that in surrounding normal tissues to ensure a high ratio (about 3:1) of uptake of boron in tumors (8). BPA conjugated with fructose has been proven to increase its solubility, so that the drug uptake in tumor is enhanced (9,10).
In this study, we used 4-borono-2-18F-fluoro-l-phenylalanine-fructose (18F-FBPA-Fr) as a scintillation probe of BPA to study the pharmacokinetics of this drug in F98 glioma-bearing Fischer 344 rats. The 10B content in tumor and normal tissues or organs of glioma-bearing rats at different time points after intravenous administration of both 18F-FBPA-Fr and BPA-Fr was determined by inductively coupled plasma mass spectroscopy (ICP-MS). The biodistribution of 18F-FBPA-Fr was determined with γ-scintillation counting of removed tissues and by microautoradiography. PET scanning of 18F-FBPA-Fr injected rats was also performed.
MATERIALS AND METHODS
Preparation of 18F-FBPA-Fr
18F-FBPA-Fr was prepared using the method as described previously by Imahori et al. with some modifications (11). Radiolabeled 18F-F2 was produced through the 20Ne(d,α)18F reaction using a Scanditronix MC17F cyclotron in the National PET and Cyclotron Center (Veterans General Hospital, Taipei, Taiwan). About 5.55 GBq 18F-F2 were produced from Ne mixed with 200 μmol of carrier F2 in an aluminum target body irradiated with 8.5-MeV deuteron for 2 h at 30-μA beam current. 18F-Acetyl hypofluorite (18F-AcOF) was generated by passing 18F-F2 through a 5.6 × 35 mm (inside diameter [ID]) cartridge containing 500 mg of tightly packed KOAc/HOAc powder (1:1.5) prepared as described by Jewett et al. (12). The effluent from the cartridge was bubbled (flow rate, 40 mL/min) into a 5-mL conical Reacti vial containing BPA (20 mg, 100 μmol; New Concept Therapeutics, Inc.) in 4-mL trifluoroacetic acid at ambient temperature. The trifluoroacetic acid was removed under reduced pressure. Acetic acid (0.1%, 2 mL) was used to dissolve the residue and then filtered through a 0.22-μm membrane filter (Millex-GV [reference no. SLGVR25LS]; Millipore Corp.). The 18F-FBPA was purified by a reverse-phase high-performance liquid chromatography (HPLC) separation system. The preparative HPLC system includes a radial compression module (RCM; Waters Corp.) containing a Delta-Pak C18 guard cartridge and column (25 × 10 mm ID length and 25 × 100 mm ID, respectively), an ultraviolet detector (Waters 486 tunable absorbance detector; Waters Corp.), and a radioactivity detector (Bioscan, Inc.). Acetic acid (0.1%) was used as the mobile phase (flow rate, 10 mL/min). The 18F-FBPA that eluted between 24 and 29 min was collected, and the radiochemical purity was determined with analytic HPLC (100-RP-18, 5 μm, 4 × 250 mm column [LiChrospher]; eluent, methanol/0.8% acetic acid containing 1 mmol/L ethylenediaminetetraacetic acid and 1 mmol/L octyl sodium sulfate [15:85, v/v]; flow rate, 1 mL/min). The solvent was removed under reduced pressure; then sodium bicarbonate (0.5 mL, 8.4%) and fructose (1.0 mL, 0.5 mol/L) were added and the solution was filtered through a 0.22-μm membrane filter (Millex-GV [reference no. SLGV013SL]; Millipore Corp.) into a sterile vial to afford the final product of 18F-FBPA-Fr. The overall synthetic time was 110 min.
F98 Glioma Brain Tumor Model in Rats
Male Fischer 344 rats (12–14 wk old, about 250–280 g) were anesthetized intraperitoneally with a mixture of ketamine and xylazine. Then 1 × 105 F98 rat glioma cells (a generous gift from Dr. Rolf F. Barth, Ohio State University) in 10 μL Hanks’ balanced salt solution without Mg2+ and Ca2+ were injected into the left brain region. F98 cells were slowly (15–20 s) injected into the brain, the syringe was held still for 2 min, and the needle was then withdrawn. The hole was sealed with bone wax. Finally, the wound was flushed with iodinated alcohol and held together with a sterilized steel clip. The animal experiments were approved by the Laboratory Animal Care Panel of National Yang Ming University.
Biodistribution of 18F-FBPA-Fr in F98 Glioma-Bearing Fischer 344 Rats
The biodistribution of 18F-FBPA-Fr was determined on the 13th day after tumor implantation. Tumor-bearing rats weighing 250–280 g, anesthetized with ether (catalog no. 100946B; Merck), were injected with 8.15–10.0 MBq 18F-FBPA-Fr through lateral tail veins. At 0.5, 1, 2, and 4 h after injection, rats were killed with chloroform (catalog no. 1.02445; Merck). Tumors, surrounding normal brain tissues, pancreas, blood, kidneys, small intestine, spleen, and liver were removed, and parts of these tissues or organs were assayed for radioactivity with a γ-scintillation counter (Cobra II Autogamma; Packard). The uptake of 18F-FBPA-Fr in tissues or organs was expressed in counts per minute (cpm) corrected with decay and normalized as the percentage injected dose per gram of tissue (%ID/g) according to the following formula:
where ln(A/A0) = −0.693t/t1/2, A = radioactivity (cpm) of tissues or organs measured by γ-counter, A0 = decay-corrected radioactivity (cpm) of tissues or organs, Eff = counting efficiency of γ-scintillation counter, t1/2 = half-life of radioisotope, and t = time after injection.
10B Assay in F98 Glioma-Bearing Fischer 344 Rats After BPA-Fr or 18F-FBPA-Fr Injection
The 10B assay was performed with the same animal model and administration route of both drugs. Test substances (172 mg/kg body weight for BPA-Fr and 8.15–10.0 MBq for 18F-FBPA-Fr) were injected via the tail vein. The 10B concentrations in tumor and tissues were determined by ICP-MS and were normalized as μg/g of tissue.
Microautoradiography
At 0.5, 1, 2, and 4 h after 18F-FBPA-Fr injection, rats were killed with chloroform, and whole brains were surgically removed, frozen immediately with dry ice, and then embedded with Tissue-Tek OCT (optimal cutting temperature) compound (catalog no. 4583; Sakura Finetechnical Co., Ltd.) on round specimen disks (diameter, 2.2 cm). The embedded samples were placed on a −30°C freezing stage in the cryostat (CM 3050; Leica) for about 30 min. The coronal sectioning was performed with a slice thickness of 15 μm. Sections attached on the microscopic slides were air dried at room temperature and applied to imaging plates (BAS cassette 2040; Fujifilm) and exposed for about 36 h. After exposure, the imaging plates were assayed with a BAS-2500 IP reader (Fuji Photo Film Co.).
Quantification of 18F-FBPA-Fr Biodistribution in Microautoradiography
After scanning the microautoradiographic sections, the images were measured by a bioimage reader connected to a Pentium IV computer where images were stored in TIF format. The specificities of the reading conditions were as follows: resolution of 50, gradation of 16 bits, dynamic range of L5, and sensitivity of 10,000. The stored images were analyzed with Adobe Photoshop 6.0. The regions of interest (ROIs) were fixed at 1,643 pixels to obtain the average optical density (OD) for each ROI measured. Repeated measurements for ANOVA were applied to find the statistical significance of the results.
PET Scanning
PET images of F98 glioma-bearing rats were obtained using an ECAT HR+ PET system (Siemens/CTI) that produced 63 image slices over a 15.52-cm axial field of view. The spatial resolution is 4.6 mm in the central axis using the 2-dimensional brain mode. PET scanning was performed on the 13th day after tumor implantation. Each tumor-bearing rat was anesthetized with halothane vapor using a vaporizer system (Fluosorber; Int. Market Supply) and injected with 20.35-MBq 18F-FBPA-Fr through a lateral tail vein. Data acquisition by PET scanning was initiated at the first minute after drug injection. Dynamic coronal images were acquired using ten 60-s frames and ten 2-min frames, followed by 10-min frames up to 4 h after injection. Transverse scanning was also performed at 4 h after injection. All images were reconstructed using the weighted attenuation method, ordered-subsets expectation maximization, with 128 × 128 pixel image size, 16 subsets, a zoom factor of 5, and use of a gaussian filter.
Time–activity curves were plotted for both tumor and reference ROIs, located in the contralateral normal brain. ROIs of tumor were drawn from each image plane in which tumor was visible on the final time frame. The radioactivities of tumor ROIs at different time frames were calculated. Reference ROIs were drawn from the final frame at the end of scanning in the contralateral normal brain and applied to all images in the dynamic sequence.
RESULTS
Preparation of 18F-FBPA-Fr
The radiochemical purity of purified 18F-FBPA was >97% as determined with HPLC (Fig. 1). The radiochemical yield of 18F-FBPA-Fr (444–518 MBq corrected to end of bombardment [EOB] based on the radioactivity of 18F-AcOF) was 20%–25% after about 10 runs.
HPLC chromatograms of 18F-FBPA solution. (A) Crude product, determined with preparative HPLC: retention time of 18F-FBPA was about 24.9 min. (B) Purified product, determined with analytic HPLC: retention time of 18F-FBPA was about 8.6 min.
Biodistribution of 18F-FBPA-Fr in F98 Glioma-Bearing Fischer 344 Rats
The uptake of 18F-FBPA-Fr in F98 glioma reached the maximum level at 1 h after injection (Figs. 2 and 3; Table 1). Although the tumor-to-heart blood ratio showed a steady 2-fold uptake during the 4-h study (1.72, 2.61, 2.00, and 1.93 at 0.5, 1, 2, and 4 h after 18F-FBPA-Fr injection), the tumor-to-normal brain ratio reached the maximum earlier (3.45 at 0.5 h after injection) and decreased slowly with time (3.13, 2.61, and 2.25 at 1, 2, and 4 h after injection). The biodistribution of 18F-FBPA-Fr for the other tissues or organs after a single injection of 18F-FBPA-Fr is also shown in Figure 2 and Table 2. The tumor uptake reached the maximum (0.47 ± 0.12 %ID/g) at 1 h after intravenous injection of 18F-FBPA-Fr and dropped to 0.27 ± 0.05 %ID/g at 4 h after injection. The kidneys had the highest radioactivity levels and pancreas had the second highest radioactivity levels up to 4 h after injection.
Time–activity curves of 18F-FBPA-Fr in various organs of F98 glioma-bearing Fischer 344 rats after intravenous injection (8.15–10.0 MBq).
10B concentration in various organs of F98 glioma-bearing Fischer 344 rats assayed by ICP-MS. Injection dose of BPA-Fr was 43 mg per rat.
Biodistribution of 18F-FBPA-Fr in Tumor, Left Brain, Right Brain, and Blood of F98 Glioma-Bearing Fischer 344 Rats at 0.5, 1, 2, and 4 Hours After Intravenous Injection of 8.15–10.0 MBq 18F-FBPA-Fr
Biodistribution of 18F-FBPA-Fr in Tumor and Other Normal Tissues of F98 Glioma-Bearing Fischer 344 Rats at 0.5, 1, 2, and 4 Hours After Intravenous Injection of 8.15–10.0 MBq 18F-FBPA-Fr
10B Assay in F98 Glioma-Bearing Fischer 344 Rats After BPA-Fr and 18F-FBPA-Fr Injection
The 10B content in F98 glioma reached the maximum at 1 h after injection of both BPA-Fr and 18F-FBPA-Fr (Fig. 3; Tables 3 and 4). The result is consistent with that of 18F-FBPA-Fr shown in Table 2. However, tumor-to-left brain ratios were highest at 0.5 h after injection in both cases (Tables 3 and 4).
10B Concentration of Tumor, Left Brain, and Right Brain in F98 Glioma-Bearing Fischer 344 Rats at 0.5, 1, 2, and 4 Hours After Intravenous Injection of BPA-Fr (172 mg/kg)
10B Concentration of Tumor, Left Brain, and Right Brain in F98 Glioma-Bearing Fischer 344 Rats at 0.5, 1, 2, and 4 Hours After Intravenous Injection of 8.15–10.0 MBq 18F-FBPA-Fr
Brain Microautoradiography
The biodistribution of 18F-FBPA-Fr imaged by microautoradiography in rats bearing F98 gliomas is shown in Figure 4. The higher OD regions (Fig. 4, top) are well correlated with the anatomic localization of the tumors (Fig. 4, bottom). The tumor regions could be detected as viewed by phosphor plate imaging at 0.5 h after injection. The quantification of relative uptake of 18F-FBPA-Fr in tumor and normal brain from microautoradiograms is shown in Figure 5. The accumulation of 18F-FBPA-Fr in tumors from 0.5 to 4 h after drug administration was significantly higher than that of surrounding normal brain tissues, in which 1 h was the highest. The results further confirm that the uptake of 18F-FBPA-Fr was higher in tumor than in surrounding normal brain tissue.
Microautoradiography of brains in F98 glioma-bearing Fischer 344 rats after intravenous injection of 54.76- MBq 18F-FBPA-Fr. Times after drug administration are indicated.
Relative optical densities of tumor and surrounding normal brain derived from microautoradiography as shown in Figure 4. **P < 0.01.
PET Scanning of 18F-FBPA-Fr in F98 Glioma-Bearing Fischer 344 Rats
Coronal views of PET images of F98 glioma-bearing Fischer 344 rats are shown in Figure 6. The glioma in the brain can be clearly demonstrated through uptake of 18F-FBPA-Fr on coronal images. Kidneys again showed the highest radioactivity levels. The time–activity curves of 18F-FBPA-Fr in both tumor and normal brain are shown in Figure 7, which demonstrated that the accumulation of radioactivity in tumor was increased rapidly at first hour and then gradually declined up to 4 h after drug injection.
PET images of F98 glioma-bearing Fischer 344 rat at 0.5, 1, 2, and 4 h after intravenous injection of 20.35 MBq- 18F-FBPA-Fr (coronal view).
Time–activity curve of tumor and normal brain tissue in F98 glioma-bearing Fischer 344 rat derived from PET images after intravenous injection of 20.35-MBq 18F-FBPA-Fr.
DISCUSSION
The methodology of BNCT is complicated in 2 respects: neutron dosimetry and the neutron-capturing efficiency of boron compounds. Each component could be manipulated independently, so that the optimal time between administration of the neutron capture agent and neutron irradiation can be adjusted for differential uptake of 10B between tumor and normal tissues. Since 10B accumulation varies with the nature of the tumor, which even with the same grades often varies in their biochemical properties, it is suggested that the 10B concentration in tumors should be determined for each individual patient before neutron irradiation. The collection of accurate data for individual patients is very important for performing BNCT and neutron dosimetry. 10B levels in tumor and surrounding normal brains in vivo play a key role in BNCT (13). Nonetheless, assay of the time–activity curve of the 10B concentration in tumor and normal tissues often is not easy to perform. A noninvasive method for the boron distribution assay would be valuable to patients who are in poor surgical condition. Imahori et al. designed a method for quantitative measurement of 10B concentration in patients with high-grade gliomas by PET imaging with 18F-FBPA (13). According to their study, the estimated values of 10B concentration in gliomas were very close to those of the 10B concentration in surgical specimens based on the incorporation constant (Ic*). This method was based solely on PET imaging and could potentially provide data that would assist in the selection of patients for the BNCT treatment after surgical resection of brain tumors. Ishiwata et al. also demonstrated that the uptake of 10B-BPA was similar to that of 18F-FBPA in pharmacokinetics. The concentration of l-BPA in tumor and normal tissues assayed by ICP-MS corresponded almost 1:1 in hamsters receiving 18F-FBPA (14).
BPA has been reported to be a safe agent in BNCT for glioma patients (15). Neither toxicity resulting from the intravenous infusion of 250–290 mg/kg of BPA-Fr nor radiation-induced tissue damage caused by neutrons has been reported. The high uptake of 18F-FBPA-Fr in the pancreas of rat was also found in this study, which is consistent with the finding in mice as previously reported by Ishiwata et al. (16). Thus, 18F-FBPA-Fr could be also applied in evaluating the function of the pancreas by PET imaging. The 10B content in glioma and normal brain after administration of BPA-Fr was parallel to that of 18F-FBPA-Fr, suggesting that the 10B content in various tissues or organs could also be estimated from the uptake of 18F-FBPA-Fr in normal tissues or organs after administration of 18F-FBPA-Fr. This may be an advantage of PET scanning as a tool in monitoring the drug distribution for optimal neutron irradiation time in BNCT.
Microautoradiography of tumors showed that the high OD regions were well correlated with the localization of gliomas. This technique also demonstrated that tumor had higher uptakes of 18F-FBPA-Fr than that of normal brain. Gliomas could also be visualized in dynamic PET imaging. The kidneys and pancreas both showed high radioactivity levels. The time–activity curves of tumor and normal brain tissues derived from dynamic PET images also showed that the radioactivity of the tumor reached the maximum at 1 h after drug injection and then decreased with the increase in time. The results were also consistent with the findings from the biodistribution study. The noninvasive PET assay of the concentration of the positron-emitted radioisotope-labeled compounds—such as 18F-FBPA-Fr in tumors and normal tissues in vivo—not only provides images of tumors but also allows determination of the concentration of the atom of interest in various tissues or organs if the calibration curve of the biodistribution study is investigated in parallel, such as the pharmacologic characteristics of 18F-FBPA-Fr and BPA-Fr.
CONCLUSION
18F-FBPA-Fr shows high tumor-to-normal tissue uptake ratio in F98 glioma-bearing rats and can be used as a probe for BPA-Fr in BNCT for the treatment of brain tumors. This study provides useful information for the future clinical application of BNCT in cancer therapy.
Acknowledgments
This study was supported by grants NSC89-2745-P-010-003, NSC91-2745-P-010-003 and NSC92-2745-P-010-002 from National Science Council, Taipei, Taiwan. We thank the staff of the National PET and Cyclotron Center in Taipei Veterans General Hospital, who kindly provided the radiopharmaceuticals. The PET Gene Probe Core of National PET/Cyclotron Center, Veterans General Hospital, Taipei, Taiwan, is also gratefully acknowledged. We also thank the staff of the Instrumentation Center at the National Tsing Hua University, Hsinchu, Taiwan, for technical support of the tissue boron concentration assay. The imaging instrument (FLA5000, Fuji Photo Film Co.) was supported by grant 89-B-FA22-1-4-05 to the National Yang-Ming University, Taipei, Taiwan, for promoting academic excellence of universities from the Ministry of Education of Taiwan. Part of this study was presented at the 10th International Congress on Neutron Capture Therapy in Essen, Germany, September 8–13, 2002.
Footnotes
Received Apr. 7, 2003; revision accepted Oct. 9, 2003.
For correspondence or reprints contact: Jeng-Jong Hwang, PhD, Department of Medical Radiation Technology, Institute of Radiological Sciences, National Yang-Ming University, 155 Li-Nong St., Section 2, Pei-tou, Taipei 112, Taiwan.
E-mail: jjhwang{at}ym.edu.tw