Comparison of Radiolabeled Nucleoside Probes (FIAU, FHBG, and FHPG) for PET Imaging of HSV1-tk Gene Expression

Cell Lines

RG2 rat glioma cell line was kindly provided by Dr. Darell Bigner (Duke University Medical Center, Durham, NC). RG2 cells were transduced with the recombinant replication-deficient STK retrovirus containing the NeoR gene and HSV1-tk gene as described (1). The transduced cell line, RG2TK+, has remained stable and has expressed constant levels of HSV1TK+ since 1994; this cell line has been characterized previously (1,2).

¹⁸F-FHBG and ¹⁸F-FHPG Syntheses

The syntheses of ¹⁸F-FHBG and ¹⁸F-FHPG were performed with minor modifications according to procedures reported in the literature (16–18). Briefly, 3.7 GBq (100 mCi) of no-carrier-added aqueous H¹⁸F from the cyclotron target was added to a solution of potassium carbonate (2–3 mg) and Kryptofix 2.2.2. (12–15 mg; Aldrich Chemical Co., Milwaukee, WI) in CH₃CN and H₂O (88:12). Azeotropic distillation at 115°C with freshly distilled acetonitrile (3 × 1 mL) under a nitrogen stream efficiently removed the target H₂O. N²-(p-Anisyldiphenylmethyl)-9-[(4-tosyl)-3-p-anisyldiphenylmethoxymethylbutyl]guanine (16–18) (3–4 mg dissolved in 0.5 mL of freshly distilled acetonitrile) was introduced to the anhydrous potassium fluoride-Kryptofix 2.2.2. complex. The reaction mixture was heated at 120°C for 20 min and was subsequently allowed to cool, at which time the crude product was passed through a silica Sep-Pak cartridge to remove Kryptofix 2.2.2. and nonreacted fluoride. The Sep-Pak was eluted with 15% methanol/dichloromethane (3.5 mL) and evaporated in vacuo. The residue was acidified with 1N HCl (0.6 mL) and heated for 10 min at 120°C. Upon cooling, the contents were neutralized with 6N NaOH (0.1 mL). The crude product was purified by high-performance liquid chromatography (HPLC) on a semipreparative column (Luna C-18, 10 × 250 mm; Alltech Associates, Inc., Deerfield, IL). The mobile phase consisted of 5% EtOH/H₂O and the flow rate was set at 5 mL/min. ¹⁸F-FHBG eluted at 12 min under these conditions. ¹⁸F-FHPG was prepared similarly from N²-(p-anisyldiphenylmethyl)-9-[(1-anisyldiphenyl)-3-tosyl-2-propoxymethyl]guanine (16–18) and was purified under the same conditions as described above; ¹⁸F-FHPG eluted at 7 min.

The radiochemical purities of ¹⁸F-FHBG and ¹⁸F-FHPG were >97% as measured by radio-HPLC using an analytic Luna C-18 column (4.5 × 250 mm) and 6% acetonitrile/H₂O as the mobile phase. For specific activity calculations, standard solutions of each compound were injected to the HPLC column, and the area under ultraviolet (UV) was measured. The radioactive product with known activity was injected to the HPLC column under the same condition, and the area under UV was measured. This area under the UV peak was converted to mass, and activity was divided by the mass to yield specific activity for that preparation (16). In our previous studies we reported the estimated lower limits of specific activities of ¹⁸F-FHBG and ¹⁸F-FHPG to be 11.8 GBq/μmol (320 Ci/mmol) (18) and 19.5 GBq/μmol (526 Ci/mmol) (16), respectively. In more recent experiments, we found that a small amount of UV active impurity elutes close to the product. We initially thought this peak represented unlabeled product; however, this was subsequently determined to be incorrect. Therefore, a more likely estimate of the specific activity for ¹⁸F-FHBG and ¹⁸F-FHPG used in these experiments is in the range of 44.4–74.0 GBq/μmol (1,200–2,000 Ci/mmol) (although this was not determined at the time the syntheses were performed for the studies reported in this article). The final product was tested for sterility and pyrogenicity by standard techniques and found to be sterile and free from pyrogens.

In Vitro Assays

RG2TK+ and RG2 cells in culture were used to compare the accumulation of ¹⁴C-FIAU (2.07 GBq/mmol [56 mCi/mmol], >99.7% radiochemical purity; Moravek Biochemicals, Brea, CA) and each of the guanosine analogs, ¹⁸F-FHBG and ¹⁸F-FHPG, as described (1). Paired time-course radiotracer accumulation experiments were run over 180 min and the data were normalized to ³H-thymidine ([TdR] 2,220 GBq/mmol [60 Ci/mmol], >99.5% radiochemical purity; Moravek Biochemicals) accumulation. The incubation medium contained ¹⁴C-FIAU (0.37 kBq/mL [0.01 μCi/mL], 0.18 μmol/L), ¹⁸F-FHBG (148 kBq/mL [4 μCi/mL], 7.6 pmol/L), or ¹⁸F-FHPG (148 kBq/mL [4 μCi/mL], 12.5 pmol/L) and ³H-TdR (3.7 kBq/mL [0.1 μCi/mL], 0.16 pmol/L). The radioactivity assay for the in vitro tissue culture studies involved ¹⁸F γ-counting (AutoGamma 5550 spectrometer; Packard Instruments, Meriden, CT) immediately after the experiment, followed 24–36 h later by ³H and ¹⁴C β-isotope counting (Tri-Carb Liquid Scintillation Analyzer, model 1600TR; Packard Instruments) using external standard quench correction and standard dual-counting techniques. All data were expressed as disintegrations per minute (dpm)/g cells (or dpm/mL medium). The net accumulation rate (K) was calculated from the slope of the probe accumulation versustime plot (dpm/g cells ÷ dpm/mL medium vs. time; units = mL medium/g cells/min − a clearance constant).

¹²⁴I-FIAU Synthesis

¹²⁴I-FIAU was synthesized as described (6) with some modifications. Briefly, ¹²⁴I was produced on the Memorial Sloan-Kettering Cancer Center (MSKCC) CS-15 cyclotron using the ¹²⁴Te(p,n)¹²⁴I nuclear reaction on an enriched ¹²⁴TeO₂/Al₂O₃ solid target (20). The 2′-fluoro-2′-deoxy-1-β-d-arabinofuranosyl-5- (tri-n-butyltin)-uracil (FTBSnAU) precursor was prepared by an adaptation of a procedure by Vaidyanathan et al. (21) in 69% yield and was identified by ¹H and ¹⁹F nuclear magnetic resonance spectroscopy and mass spectrometry. Na¹²⁴I was added to a solution of FTBSnAU (125 μg) dissolved in 50 μL methanol, followed by the addition of a 20-μL mixture of 30% hydrogen peroxide/acetic acid (1:3, v/v). Aqueous saturated sodium metabisulfate (0.1 mL) was added to quench the reaction; then the reaction mixture was loaded onto a C₁₈ Sep-Pak cartridge system (Waters, Milford, MA). The C₁₈ cartridge system was eluted with water (35 mL), followed by methanol (3–4 mL) to isolate the ¹²⁴I-FIAU. The methanol was evaporated and the ¹²⁴I-FIAU was formulated in physiologic saline (with 5% ethanol added), which was passed through a sterile 0.22-μm Millipore filter (Millipore, Inc., Bedford, MA).

The radiochemical purity was determined by radio-thin-layer chromatography (silica gel plates [Sigma-Aldrich, St. Louis, MO]; eluent, ethyl acetate/acetone/H₂O, 14:8:1). ¹²⁴I-FIAU comigrated with authentic FIAU standard. The radiolabeled product was isolated in 51%–54% radiochemical yield (3 labelings), with radiochemical purities of ≥97%. The specific activity of the initial Na¹²⁴I was found to be >1,110 GBq/μmol (>30 Ci/μmol) (20), so that the specific activity of the ¹²⁴I-FIAU was estimated to be similar to this value. The structure of FIAU was confirmed by single-crystal x-ray analysis.

¹³¹I-FIAU Synthesis

No-carrier-added ¹³¹I-FIAU was prepared by the same iododestannylation method (21) as was used for ¹²⁴I-FIAU. Commercially produced Na¹³¹I was obtained from New England Nuclear (North Billerica, MA). Radiochemical yields obtained with ¹³¹I averaged 93% (3 labelings), and radiochemical purities were found to be ≥99%. The specific activities of the ¹³¹I-FIAU produced were estimated to be 48–59 GBq/μmol (1.3–1.6 Ci/μmol) on the basis of the manufacturer’s specifications accompanying the Na¹³¹I lots used.

Subcutaneous Xenografts and Animal Studies

The experimental protocol involving animals was approved by the Institutional Animal Care and Use Committee of MSKCC. Subcutaneous xenografts were produced in rnu/rnu rats (Frederick Cancer Center, Frederick, MD) weighing 200–250 g by subcutaneous injection of 5 × 10⁶ tumor cells in 200 μL of serum-free cell culture medium under anesthesia (ketamine [87 mg/kglrsqb] and xylazine [13 mg/kg], intraperitoneally). Transduced RG2TK+ cells were placed in the left shoulder (test) and wild-type RG2 cells were placed in the right shoulder (negative control). The animals were monitored for tumor growth by daily measurements of the tumor’s size and the animal’s weight. Animals were studied when the subcutaneous tumor xenografts reached a diameter of ∼15 mm, 7–14 d after subcutaneous implantation of the RG2 and RG2TK+ cells. The in vivo growth rate, macroscopic and microscopic appearances, and the degree of vascularization were similar in the HSV1-tk transduced and wild-type subcutaneous xenografts.

Three days before study, all animals received 1.0-mL intravenous injections of a 0.9% NaI solution (30 mg/kg NaI) daily for 3 consecutive days to block thyroid uptake of radioactive iodide. For the paired 2-h FHBG-24-h FIAU imaging studies (n = 6), the animals were first injected with ¹⁸F-FHBG (27.8 MBq [0.75 mCi] per animal intravenously), imaged over 2 h, and then injected with ¹²⁴I-FIAU (8.9 MBq [0.24 mCi] per animal intravenously) and reimaged 24 h later. For the paired 2-h FHPG-24-h FIAU imaging studies (n = 6), the animals were first injected with ¹⁸F-FHPG (61.1 MBq [1.65 mCi] per animal intravenously), imaged over 2 h, and then injected with ¹²⁴I-FIAU (13.0 MBq [0.35 mCi] per animal intravenously) and reimaged 24 h later. For the paired 2-h FHBG-2-h FIAU imaging studies (n = 6), the animals were first injected with ¹⁸F-FHBG (50.0 MBq [1.35 mCi] per animal intravenously) and imaged over 2 h; 24 h later the same animals were injected with ¹²⁴I-FIAU (7.0 MBq [0.19 mCi] per animal intravenously) and reimaged over 2 h. Immediately after the ¹²⁴I-FIAU imaging sessions, animals were euthanized and tissue samples were processed for ¹²⁴I radioactivity using an AutoGamma 5550 spectrometer (Packard Instruments). Radioactivity is expressed as percentage dose/g tissue (%dose/g tissue).

For the paired 2-h ¹³¹I-FIAU-¹⁸F-FHBG and the 2-h ¹³¹I-FIAU-¹⁸F-FHPG tissue sampling studies, 2 groups of animals were studied. Simultaneous intravenous injection of ¹³¹I-FIAU (1.81 MBq [0.049 mCi] per animal) and ¹⁸F-FHPG (14.8–25.9 MBq [0.40–0.70 mCi] per animal) was performed in 1 group (n = 6), and injection of ¹³¹I-FIAU (3.18 MBq [0.086 mCi] per animal) and ¹⁸F-FHBG (5.55–13.0 MBq [0.15–0.35 mCi] per animal) was performed in the other group (n = 6). All animals were euthanized at 2 h after injection, and the tissue samples were processed for ¹³¹I and ¹⁸F radioactivity assay using an AutoGamma 5550 spectrometer. The upward cross-talk of ¹³¹I into the ¹⁸F channel was 2.7% of the counts in the ¹³¹I channel. Corrections for ¹⁸F and ¹³¹I decay and ¹³¹I upward cross-talk were applied to all samples. Radioactivity in the animal studies is expressed as %dose/g tissue.

PET

PET imaging was performed using the Advance Tomograph (General Electric Medical Systems, Milwaukee, WI), with a spatial resolution of 5-mm full width at half maximum (FWHM) at the center of the field of view. This camera was cross-calibrated previously with the AutoGamma 5550 spectrometer. All animals were anesthetized with ketamine (87 mg/kg) and xylazine (13 mg/kg) administered intraperitoneally. Measured attenuation correction was performed using a 7-min-duration transmission scan with two 107.3-MBq (9 mCi) germanium transmission sources. Two-dimensional emission scans were obtained for all studies.

Four studies (n = 6 animals per study) acquired dynamic image data from 10 or 20 min to 130 min after injection; acquisition frames were either 10 or 20 min in duration, depending on the dose of radiopharmaceutical administered. Two studies (n = 6 animals per study) involved imaging at 24 h after injection of ¹²⁴I-FIAU (20-min acquisition).

Emission counts were corrected for random coincidences, dead time, and scatter. Emission scans were reconstructed using an iterative reconstruction method with measured attenuation correction, smoothed with an 8-mm gaussian filter. The reconstruction parameters were 28 subsets, 2 iterations in a 256 × 256 matrix using a loop filter of 2.15-mm FWHM and a postfilter of 3.0-mm FWHM. Regional tumor radioactivity concentrations (%dose/mL) were estimated from the maximum pixels within regions of interest drawn around the tumor on transaxial slices of the reconstructed image sets.

Statistical Analysis

The statistical significance of comparisons between 2-h FIAU and either 2-h FHBG or 2-h FHPG values for RG2-TK+ and the ratios of RG2-TK+ to RG2 or muscle values was based on 2-sided paired Student t tests for matched values in the same rat. The statistical significance of comparisons between 2-h FIAU and 24-h FIAU values was based on 2-sided 2-sample t tests. Results are reported as mean ± SD. Logarithmic transformations brought the data into closer agreement with implicit assumptions of normal statistical distributions with constant variance.

In Vitro Assays

A paired comparison of ¹⁴C-FIAU, ¹⁸F-FHBG, and ¹⁸F-FHPG accumulation, normalized to ³H-TdR in transduced RG2TK+ cells, is shown in Figure 1. The slopes of the plots show a 17-fold greater uptake of FIAU compared with that of FHBG and a 41-fold greater uptake of FIAU compared with that of FHPG. The mean values (±SD) of the tissue culture medium clearance constant (K) calculated for RG2TK+ cells were 0.38 ± 0.05, 0.023 ± 0.001, and 0.0077 ± 0.0004 mL/min/g cells for FIAU, FHBG, and FHPG, respectively. The mean K values for wild-type RG2 cells (data not shown) were much lower: 0.076 ± 0.013, 0.0032 ± 0.0026, and 0.0004 ± 0.0003 mL/min/g cells for FIAU, FHBG, and FHPG, respectively.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Paired nucleoside accumulation studies in RG2TK+ transduced cells. (A) ¹⁸F-FHBG-¹⁴C-FIAU (⊞) and ¹⁸F-FHPG-¹⁴C-FIAU (○) accumulation data (mL medium/g cells) are shown. (B) ¹⁸F-FHBG-¹⁴C-FIAU and ¹⁸F-FHPG-¹⁴C-FIAU comparisons with ³H-TdR are shown: ¹⁴C-FIAU (•), ¹⁸F-FHBG (⊞), and ¹⁸F-FHPG (□). Each point represents paired comparison of uptake results obtained from single triple-label experiment; experimental times varied between 10 and 180 min.

In Vivo Imaging

A comparison of ¹²⁴I-FIAU, ¹⁸F-FHBG, and ¹⁸F-FHPG PET images is shown in Figure 2. Each quadrant shows a pair of images obtained from the same animal on sequential days. The 24-h FIAU images (animals 1 and 3) show localization of high levels of ¹²⁴I radioactivity in the transduced RG2TK+ xenograft and background levels of radioactivity in the contralateral RG2 wild-type xenograft (compared with shoulder muscles and thorax). Modest levels of radioactivity are also seen in the stomach, and lower levels of radioactivity are seen in the intestine. The corresponding 2-h FHBG (animal 1) images show low ¹⁸F activity in the transduced RG2TK+ xenograft and background levels of radioactivity in the RG2 wild-type xenograft. The corresponding 2-h FHPG (animal 3) images show near-background ¹⁸F activity in the transduced RG2TK+ xenograft and background levels of radioactivity in the RG2 wild-type xenograft. Xenograft localization to the shoulders was confirmed by the corresponding transmission scan images (e.g., animal 3 in Fig. 2). Background (thorax) radioactivity (%dose/g) was similar on the 2-h FHBG and FHPG images and on the 24-h FIAU images.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Paired PET imaging studies comparing ¹²⁴I-FIAU and ¹⁸F-FHBG or ¹⁸F-FHPG were performed. Coronal (A) and axial (B) PET images are shown for 3 different rats. Each animal had 2 imaging studies within 24 h, and comparable image pairs are shown in each quadrant of A and B. Location of RG2TK+ xenograft is indicated by arrow. Linear scaling of image intensity was performed to achieve similar background intensity on all images; background radioactivity in thorax was 0.013 and 0.027 %dose/g for animal 1, 0.024 and 0.18 %dose/g for animal 2, and 0.23 and 0.24 %dose/g for animal 3 (FHBG or FHPG and FIAU, respectively). No radioactivity above background is visualized in control RG2 xenograft. Several tissues had substantially higher values than indicated on intensity scale. For example, FIAU levels in RG2TK+ xenografts in animals 1–3 were 0.55, 1.7, and 1.2 %dose/g, respectively. Also note high values of radioactivity in stomach (6.0 %dose/g) and bladder (3.9 %dose/g) in 2-h study (animal 2). FHBG levels in abdominal viscera (2.8–3.2 %dose/g) and bladder (11 and 80 %dose/g) were also substantial at 2 h (animals 1 and 2); similar values were measured for FHPG in abdomen (3.4 %dose/g) and bladder (63 %dose/g) of animal 3. Note that FIAU bladder radioactivity had essentially cleared by 24 h (animals 1 and 3).

The comparison of a 2-h FHBG image with a corresponding 2-h FIAU image from animal 2 is shown in Figure 2. The FIAU radioactivity levels in the RG2TK+ xenografts at 2 and 24 h were similar (animals 1–3). The level of radioactivity in the RG2 xenograft was at background levels (e.g., shoulder muscles and thorax), although background radioactivity in nontransduced tissues was significantly higher at 2 h compared with that at 24 h (Table 1). Note the higher stomach radioactivity on the 2-h image (animal 2) compared with that at 24 h (animals 1 and 3). The corresponding 2-h FHBG images of animal 2 were similar to that observed in animal 1: low ¹⁸F activity in the transduced RG2TK+ xenograft and background levels of radioactivity in the RG2 wild-type xenograft. However, gallbladder and intestine radioactivity on the 2-h FHBG images (animals 1 and 2) were substantially higher than that on the 2-h FIAU image (animal 2). Gallbladder and intestine radioactivity on the 2-h FHPG image (animal 3) was similar to that observed with FHBG (animals 1 and 2). Urinary bladder radioactivity at 2 h was high for all 3 probes; however, significant bladder radioactivity was not observed on the 24-h FIAU images (animals 1 and 3). Background (shoulder muscles and thorax) radioactivity (%dose/g) on the 2-h FIAU images (animal 2, Fig. 2) was substantially greater than that on the 2-h FHBG and FHPG images.

In Vivo Radioactivity Measurements

A comparison of FIAU, FHBG, and FHPG radioactivity (%dose/g) levels in different tissues is shown in Table 1. The xenograft tissue sampling data confirm the PET imaging findings; significantly higher levels of FIAU accumulate in RG2TK+ xenografts at 2 and 24 h compared with that of FHBG and FHPG at 2 h, and the differences between RG2TK+ and RG2 xenografts are highly significant. The difference in radioactivity between transduced RG2TK+ and wild-type RG2 xenografts and the paired RG2TK+/RG2 and RG2TK+/muscle concentration ratios are shown in Table 2. These differences and ratios were significantly greater for FIAU than for FHBG or FHPG. The FIAU radioactivity levels in RG2TK+ xenografts were similar in the 2- and 24-h experiments. However, the FIAU RG2TK+/RG2 and RG2TK+/muscle concentration ratios were significantly higher at 24 h compared with those at 2 h because of the greater washout of background radioactivity over 24 h.

FIAU/FHBG and FIAU/FHPG concentration ratio comparisons in RG2TK+ and RG2 xenografts are shown in Table 3. In transduced RG2TK+ xenografts, FIAU accumulation was 20- and 60-fold higher than that of FHBG and FHPG, respectively, for the 2- and 24-h FIAU comparisons. In nontransduced, wild-type RG2 xenografts, the FIAU/FHBG and FIAU/FHPG ratios were 10-fold for the paired 2-h experiments; this was caused by the slower clearance of FIAU compared with that of FHBG and FHPG (Table 1). A comparison of the 24-h FIAU experiments with the 2-h FHBG and FHPG data for RG2 xenografts shows much smaller differences and a low concentration ratio, again indicating the slower body clearance of FIAU.

Sequential Imaging (0–120 Minutes)

A comparison of 10-min imaging frames of ¹²⁴I-FIAU accumulation in a coronal section of a rat is shown in Figure 3. The sequence of images clearly reveals significant RG2TK+ activity above background in the initial 10-min frame. Although total body background is relatively high in the 10-min frame, it drops fairly rapidly with predominantly renal clearance. Also note that stomach activity builds rapidly and is greater than that in the RG2TK+ xenograft. The ¹²⁴I-FIAU accumulation profiles in RG2TK+ and RG2 xenografts are shown in Figure 4A. The corresponding accumulation profiles of ¹⁸F-FHBG from the same set of animals are shown in Figure 4B, and the accumulation profiles of ¹⁸F-FHPG from a different set of animals are shown in Figure 4C. FIAU accumulates rapidly and reaches a plateau in RG2TK+ xenografts during the 2-h period of imaging, whereas FIAU is being cleared from the RG2 xenograft. In contrast, FHBG and FHPG are being cleared from RG2TK+ transduced and wild-type xenografts over the 2-h imaging period. FHBG clearance from RG2TK+ xenografts appears to approach a plateau, whereas FHBG clearance from the RG2 xenografts and FHPG clearance from both xenografts fits a single exponential (there was no significant difference between the FHPG profiles in RG2TK+ and RG2 xenografts).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Comparison of 10-min imaging frames of FIAU accumulation. Ten-minute coronal image frames through RG2TK+ and RG2 xenografts are shown at various times after intravenous injection of ¹²⁴I-FIAU. Image intensity scale is same for all frames.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Radioactivity-time profiles in RG2TK+ and RG2 xenografts after intravenous injection of ¹²⁴I-FIAU, ¹⁸F-FHBG, and ¹⁸F-FHPG. Values (%dose/mL) are the mean ± SD (n = 6) of maximum pixel value measured in each xenograft. FHBG and FIAU data were obtained from same set of animals imaged on consecutive days.

The activity-time profiles of ¹²⁴I radioactivity (FIAU derived) in stomach and ¹⁸F radioactivity (FHPG derived) in intestine were assessed (data not shown). In 5 of 6 animals, ¹²⁴I radioactivity remained in the stomach; in 1 animal radioactivity was observed to move from stomach into the intestines during the 2-h imaging period. In all FHPG animals, the intestines contained the bulk of the ¹⁸F radioactivity; in 1 FHPG animal the kidneys were clearly observed. The hot spot of intestinal activity was observed to move from proximal (duodenal) to distal (large intestine) portions of the intestine during the 2-h period of imaging.