Immuno-PET and Targeted α-Therapy Using Anti–Glypican-1 Antibody Labeled with ⁸⁹Zr or ²¹¹At: A Theranostic Approach for Pancreatic Ductal Adenocarcinoma

Synthesis of [⁸⁹Zr]GPC1 mAb

⁸⁹Zr was produced via the nuclear reaction of ⁸⁹Y(p, n)⁸⁹Zr using a medical cyclotron, Cypris HM-18 (Sumitomo Heavy Industries). Mouse GPC1 mAb clone01a033 and GPC1 mAb clone T2 (a humanized antibody of clone 01a033) were generated as previously described (5,9). [⁸⁹Zr]GPC1 mAb was prepared by labeling deferoxamine-linked GPC1 mAb with ⁸⁹Zr according to previously reported methods (19). GPC1 mAb (2–2.5 mg) was dissolved in 1 mL of 0.1 M sodium bicarbonate buffer (pH 9.0) and reacted with 4–6.5 μL of 10 mM p-SCN-Bn-deferoxamine (Macrocyclics) in dimethyl sulfoxide using ThermoMixer C (Eppendorf) for 30 min (at 37°C and 550 rpm). The GPC1 mAb conjugate with deferoxamine (2 mL of solution) was purified from the reaction solution using a PD-10 column (Cytiva) and 5 mg/mL gentisic acid in 0.25 M sodium acetate buffer (pH 5.5). Next, 0.05 mL of 2 M Na₂CO₃ was added to 0.3 mL of ⁸⁹Zr solution (70–124 MBq) and incubated for 3 min. Thereafter, 0.2 mL of 0.5 M N-(2-hydroxyethyl)piperazine-N′-(2-ethanesulfonic acid) buffer (pH 7.2) and 0.6 mL of GPC1 mAb conjugate with deferoxamine solution were added and the pH was adjusted to 6.8 using 2 M Na₂CO₃, followed by a reaction at 27°C and 550 rpm for 60 min. [⁸⁹Zr]GPC1 mAb was obtained with a radiochemical yield of 17% on average and more than 90% radiochemical purity via radio–thin-layer chromatography analysis by purifying the reaction solution using a PD-10 column and 5 mg/mL gentisic acid in 0.25 M sodium acetate buffer (pH 5.5).

Synthesis of [²¹¹At]GPC1 mAb

²¹¹At was produced via the nuclear reaction of ²⁰⁹Bi(α, 2n)²¹¹At using azimuthally varying field cyclotrons (Research Center for Nuclear Physics, Osaka University, and Radioactive Isotope Beam Factory, Institute of Physical and Chemical Research) and purified using a dry distillation method (17). The isolated ²¹¹At was introduced into previously prepared [B]₁₀-GPC1 mAbs (clone 01a033 or clone 1-12). The synthesis of [B]₁₀-GPC1 mAb is described in the supplemental materials (supplemental materials are available at http://jnm.snmjournals.org). The ²¹¹At solution (clone 01a033, 13.3 MBq; clone 1-12, 12.0 MBq) and phosphate-buffered saline were added to a solution of chloramine-T in distilled water (0.5 mg/mL). Next, [B]₁₀-GPC1 mAbs (40 μg each) in phosphate-buffered saline were added to this solution and incubated at room temperature for 10 min. This reaction was quenched using Na₂S₂O₅ (0.5 mg/mL in 6 μL of distilled water). The solution was then transferred to a filter column (Cosmospin Filter H; Nacalai Tesque) packed with chromatographic resin (Sephadex G-25 DNA grade; Cytiva). The solution was centrifuged at 1,300 g for 2 min at room temperature to obtain [²¹¹At]GPC1 mAb. Radiation was measured using a germanium semiconductor detector (BE2020; Mirion Technologies). The radiochemical yields of [²¹¹At]GPC1 mAbs were calculated at 59.8% and 45.7%, respectively. The purity of [²¹¹At]GPC1 mAb clone 01a033 and [²¹¹At]GPC1 mAb clone 1-12 was checked via thin-layer chromatography (with methanol as the eluent) using a radioimager (Typhoon FLA 7000; Cytiva), and the radiochemical purity was more than 90%. An average of 2.3 decaborates was estimated to be bound to GPC1 mAb using matrix-assisted laser desorption/ionization time of flight mass spectrometry.

Evaluation of the Internalization Ability of GPC1 mAb

An internalization assay was performed using GPC1 mAbs with fluorescence conjugation in PANC-1 cells in the presence of dynasore. Dynasore is a drug that noncompetitively inhibits the guanosine triphosphatase activity of Dynamin1, a guanosine triphosphatase protein responsible for membrane fission during endocytosis. Detailed methods are described in the supplemental materials.

Preparation of Xenograft Models

PANC-1 and GPC1-positive (BxPC-3) cells were obtained from the American Type Culture Collection and European Collection of Authenticated Cell Cultures, respectively. GPC1-negative (BxPC-3 GPC1-knockout) cells were established as previously described (6). The cells were maintained in culture medium (RPMI 1640 medium with l-glutamine and phenol red [Fujifilm Wako Pure Chemical] for both PANC-1 and BxPC-3 or BxPC-3 GPC1-knockout cells) with 10% heat-inactivated fetal bovine serum and 1% penicillin–streptomycin.

Male nude mice and female nonobese diabetic/severe combined immunodeficiency mice were purchased from Japan SLC Inc. and Charles River Japan Inc., respectively. Tumor xenograft models were established by subcutaneous injection of PANC-1 (1 × 10⁷ cells in nude mice) or BxPC-3 (5 × 10⁶ cells in nonobese diabetic/severe combined immunodeficiency mice) suspended in phosphate-buffered saline (0.1 mL) and Matrigel (1:1; BD Biosciences). The mice were assessed 3 wk after implantation when the tumor size reached approximately 1 cm in diameter. The protocol was approved by the Animal Care and Use Committee of the Osaka University Graduate School of Medicine (approval number 30-088-009). Euthanasia was performed under deep anesthesia by isoflurane inhalation when signs of intolerable suffering or a significant decrease in body weight were observed.

[⁸⁹Zr]GPC1 mAb PET Imaging and Analysis

[⁸⁹Zr]GPC1 mAb (0.91 ± 0.10 MBq, ∼27 ± 3.0 μg of mouse GPC1 mAb, clone 01a033) was intravenously administered to PANC-1 xenograft mice (9 wk old; body weight, 23.4 ± 0.9 g; n = 6). PET/CT scanning was performed 1 h, 1 d, 2 d, 4 d, and 7 d after administration under isoflurane anesthesia using a small-animal PET scanner (Siemens Inveon PET/CT). After the PET scan, the mice were euthanized, and the radioactivity and weight of the major organs were determined using a well counter (BeWell; Molecular Imaging Laboratory). A comparison of tumoral uptake between GPC1-positive and GPC1-negative xenografts was performed to confirm the specificity of [⁸⁹Zr]GPC1 mAb accumulation for GPC1. [⁸⁹Zr]GPC1 mAb (1.34 ± 0.14 MBq, 34 ± 3.6 μg of humanized GPC1 mAb, clone T2) was intravenously administered to BxPC-3 and BxPC-3 GPC1-knockout xenograft mice (9 wk old; body weight, 17.3 ± 1.6 g; each n = 3). PET/CT scanning was performed 24 h after the administration of [⁸⁹Zr]GPC1 mAb, followed by the measurement of biodistribution. The blocking study was also performed by injection of a nonradiolabeled GPC1 mAb (300 μg) before injection of [⁸⁹Zr]GPC1 mAb (0.3–34 μg of humanized GPC1 mAb, clone T2). Biodistribution was compared among the following 4 groups of the PANC-1 xenograft model (each n = 3) 24 h after the administration of [⁸⁹Zr]GPC1 mAb: 20 kBq (24 ± 3 kBq, 0.36 ± 0.04 μg), 200 kBq (199 ± 14 kBq, 3.01 ± 0.21 μg), 2 MBq (2.29 ± 0.05 MBq, 34.8 ± 0.78 μg), and 2 MBq with blocking (2.26 ± 0.05 MBq, 34.2 ± 0.79 μg). PET/CT scanning was performed 24 h after the administration of [⁸⁹Zr]GPC1 mAb except in the 20-kBq group.

All PET data were reconstructed using 3-dimensional ordered-subset expectation maximization (16 subsets and 2 iterations), followed by maximum a posteriori estimation with scatter and attenuation correction. The regional uptake of radioactivity was decay-corrected to the injection time and expressed as SUV. Ellipsoid sphere regions of interest were manually placed on the tumor, heart, lungs, liver, kidneys, spleen, and bone joints using PMOD (version 3.6; PMOD Technologies).

Targeted α-Therapy Using [²¹¹At]GPC1 mAb

DNA double-strand breaks were evaluated using the γH2AX antibody as previously described (20). PANC-1 cells were seeded in 35-mm glass-based dishes and treated with 100 μM dynasore for 30 min or left untreated. [²¹¹At]GPC1 mAb (149.58 kBq) was added to these cells, which were then incubated for 120 min. After washing, Hoechst 33342 was added, and the cells were incubated for 10 min. The cells were washed and fixed with 4% paraformaldehyde phosphate buffer and then permeabilized and antibody-stained with anti-γH2AX antibody conjugated with Alexa Fluor 594 (ab206898; Abcam). A fluorescence microscope (BZ-X810; Keyence) with a 20× lens (CFI S Plan Fluor 20XC, numeric aperture 0.45; Nikon) was used for imaging.

PANC-1 xenograft mice (10 wk old; body weight, 18.4 ± 1.9 g; n = 20) were divided into 4 groups receiving intravenous administration of [²¹¹At]GPC1 mAb clone 01a033 (102.80 ± 9.87 kBq, 1.94 ± 0.19 μg, n = 5), nonradiolabeled GPC1 mAb clone 01a033 (2.08 ± 0.17 μg, n = 5), [²¹¹At]GPC1 mAb clone 1-12 (105.20 ± 12.19 kBq, 1.72 ± 0.20 μg, n = 5), and nonradiolabeled GPC1 mAb clone 1-12 (1.98 ± 0.21 μg, n = 5). Tumor size (mm³) and body weight (g) were monitored to evaluate treatment effects and systemic side effects. The therapeutic effect of internalization inhibition was also evaluated by administering the endocytosis inhibitor prochlorperazine (21). Detailed methods for immunohistochemistry are described in the supplemental materials.

Statistical Analyses

Comparisons between the 2 groups were performed using the Mann–Whitney U test or t test in SPSS (version 25.0; IBM Corp.). Differences were considered statistically significant if the P value was less than 0.05.