First-in-Human Phase 0 Study of AB001, a Prostate-Specific Membrane Antigen-Targeted 212Pb Radioligand, in Patients with Metastatic Castration-Resistant Prostate Cancer

Abstract

AB001, a prostate-specific membrane antigen (PSMA)-targeted small molecule labeled with the in vivo-generating α-emitter ²¹²Pb, was investigated in a phase 0 trial in patients with metastatic castration-resistant prostate cancer (mCRPC). The primary objective was to explore the feasibility of γ-camera imaging to assess biodistribution and uptake in metastatic lesions. Methods: Three patients with progressive mCRPC and Eastern Cooperative Oncology Group performance status 1 were included, having prostate-specific antigen levels of 0.44, 0.75, and 15 µg/L. All had at least 3 PSMA-expressing metastatic lesions, with an SUV_max range of 10.1-77.4 on PSMA PET. Each patient received a microdose of 9.4 ± 0.3 MBq of AB001 intravenously. Planar γ-camera and SPECT/CT imaging was scheduled 1-3 h and 16-24 h after administration. Whole-body clearance was assessed with NaI probe measurements. Activity of ²¹²Pb in whole blood and plasma was measured to investigate clearance from blood and in vivo stability of the ligand. Safety, tolerability, and efficacy biomarkers (prostate-specific antigen, alkaline phosphatase) were followed for 28 d. Results: AB001 uptake in the lesion with the highest PSMA expression, a retrocaval lymph node metastasis with a short-axis diameter of 11 mm, was visualized on SPECT. Uptake of AB001 was not clearly demonstrated for other metastatic lesions, possibly because of the lower PSMA expression of these metastases on PSMA PET, combined with the administered AB001 microdose and imaging system limitations. Kidney, urinary bladder with contents, and liver uptake of AB001 were clearly distinguishable from adjacent tissue, and the blood pool content was seen. Salivary glands were not visualized. Blood analyses indicated stability of AB001 after injection, and whole-body probe measurements demonstrated an effective half-life of 8 h. There were no complications related to injection of AB001 or adverse reactions during follow-up. As expected for a phase 0 study, there was no indication of therapeutic effects as assessed by prostate-specific antigen and alkaline phosphatase. Conclusion: The ²¹²Pb-based radioligand AB001 was safely administered to mCRPC patients. γ-camera imaging of AB001 was feasible, even at a microdose, and demonstrated metastatic targeting, albeit for only 1 lesion. The promising biodistribution and clearance encourage further clinical investigation.

Keywords: 212Pb; PSMA; imaging; mCRPC; targeted α-therapy; α-radioligand therapy.

Graphical abstract

FIGURE 1.

Decay scheme for ²¹²Pb, including photon emissions with energies higher than 70 keV and intensities larger than 3% (all photon intensities given per ²¹²Pb decay). ²¹²Pb emits 1 α-particle per decay via its daughter ²¹²Bi. Most imageable photons are emitted in decay to ²¹²Bi. High-energy emission of 2.6 MeV from ²⁰⁸Tl has intensity of 36% per ²¹²Pb decay and causes significant scatter that can degrade ²¹²Pb images. (Data reference: NuDat [National Nuclear Data Center, Brookhaven National Laboratory, United States].)

FIGURE 2.

SPECT and planar γ-camera imaging of AB001: biodistribution compared with baseline ¹⁸F-PSMA PET for patients ID01 (A), ID02 (B), and ID03 (C). SPECT images show maximum-intensity projections with scaling of 0–maximum for each patient. PET images show maximum-intensity projections with scaling of 0–20 SUV_max. WB γ-camera images are summed with scaling of 0–maximum.

FIGURE 3.

In patient ID03, retrocaval lymph node tumor metastasis (arrow) measuring 11 × 14 × 16 mm on CT (G and H), and with SUV_max of 77 on ¹⁸F-PSMA PET (A, B, and C). AB001 uptake was demonstrated by ²¹²Pb SPECT imaging at 4 h after administration (D, E, and F). Lymph node tumor metastasis is seen in coronal slices (A, D, and G) and axial slices (B, C, E, F, and H). PET images are scaled at 0–20 SUV_max, and SPECT images are scaled at 0%–150% of maximum of processed images (scale, 0–42 counts). Frame scatter artifact was removed from ²¹²Pb SPECT images during postprocessing.

FIGURE 4.

(A) WB activity as measured by NaI WB probe, after normalization against administered activity, for 3 patients. Monoexponential curve fit for each patient is included (solid lines). ²¹²Pb activity pharmacokinetic (PK) measurements are shown from whole blood (B) and plasma (C).

FIGURE 5.

(A) Estimated activity concentrations of ²¹²Pb for patient structures overlaid on phantom-derived contrast map for small volumes and low activity concentrations of ²¹²Pb. Background color scale illustrates contrast as measured in spheric inserts in phantom with activity concentrations down to 2 kBq/mL, imaging ²¹²Pb with SPECT reconstruction settings as in clinical study. Contrast can be seen to increase with activity concentration and volume. For patient SPECT images, some small volumes were not visualized, such as tumors of patient ID01 and salivary glands in all patients, whereas small tumor of patient ID03 was visualized. Examples of these volumes have been added to panel. Activity concentration of ²¹²Pb was estimated from PSMA PET SUV_mean, decay-corrected injected activity of ²¹²Pb, and patient body weight. Although there are uncertainties in these estimates of radioligand uptake, primarily related to imaging time points and chemical differences in PET tracer and AB001, relative positions of structures should roughly reflect probability for detection. Visual assessments in combination with phantom scan results can therefore indicate that tumor–to–parotid gland uptake ratio of AB001 is higher than that of PSMA PET tracer. Methods are described in Supplemental File 3. (B) Examples of phantom data. Background color scale in (A) was constructed from these data. Top image shows central axial slice of phantom image with activity concentration of 13 kBq/mL, and bottom image is example with 2 kBq/mL. Middle image is CT of phantom setup. (C) Baseline [¹⁸F]F-PSMA PET/CT scan from which example structures in panel were extracted for patient ID03. (D) Background-subtracted maximum-intensity projection (anterior view) of patient ID03 from SPECT imaging after AB001, with axial sections through (E) cranium at level of parotid glands and (F) lymph node metastasis. All patient SPECT images are scaled with 0%–150% of maximum count in lymph node metastasis.