Molecular Imaging for Localized Prostate Cancer: Analysis of 2 Prospective Phase 0 Trials of PSMA Targeted Immuno-PET and –SPECT, and Implications for the Radiation Oncologist

Purpose/Objective(s): To report the efficacy and utility of PSMA-based molecular targeted imaging in localized prostate cancer using ¹¹¹In-J591 SPECT and ⁸⁹Zr-J591 immuno-PET.

Materials/Methods: Initiated in 2010, two sequential phase 0 prospective studies were performed on a total of 19 patients evaluating the roles of ¹¹¹In-J591 SPECT and ⁸⁹Zr-J591 immuno-PET. All patients had a baseline MRI followed by a preoperative ¹¹¹In-J591 SPECT or ⁸⁹Zr-J591 PET scan. Patients subsequently underwent radical prostatectomy with immediate imaging of the ex vivo specimens with micro-SPECT or micro-PET/CT. For the ⁸⁹Zr-J591 trial, the ex vivo specimens were also imaged with a custom MRI solenoid coil insert for higher resolution imaging. Findings on MRI and histopathology from the surgical specimen were compared to the pre-surgical molecular imaging and ex vivo images.

Results: The ¹¹¹In-J591 SPECT: Eight patients were enrolled with a median age of 64 years. Median pre-treatment PSA was 6.8 ng/mL (2.8-22.7), and the median biopsy Gleason score was 7 (6-9). As SPECT imaging has limited lesion discriminatory power of detection, prostatectomy specimens were analyzed by the dominant tumor foci. The median pathologic Gleason score was 7 (7-9), and the median dominant lesion size was 9.5 mm (3-19 mm). Tumor was visible by MRI on 7 of the 8 patients, and was positive in all SPECT scans that correlated well to the prostatectomy specimen. One patient had positive lymph nodes on histopathology, but was negative by both MRI and SPECT. ⁸⁹Zr-J591 PET: Eleven patients were enrolled with a median age of 61.0 years. Median pre-treatment PSA was 5.2 ng/mL (3.5-12.0), and the median biopsy Gleason score was 7 (7-9). Twenty-two lesions were identified on histopathology. The median lesion size was 5.5 mm (2-21 mm), and the median pathologic Gleason score was 7 (6-9). Immuno-PET identification improved with increasing tumor size, and was true for both in vivo PET (p<0.0001) and micro-PET (p<0.0001). Furthermore, immuno-PET lesion identification improved with increasing Gleason score (for ex vivo micro-PET imaging, p=0.01; in vivo imaging, p=0.14). Lesions of Gleason score 3+3=6 were uniformly not detected by PET (0 of 14 lesions).

Conclusions: Targeted molecular PSMA imaging using the monoclonal antibody J591 has the ability to detect clinically significant localized prostate cancer. The ⁸⁹Zr-J591 PET has improved accuracy over ¹¹¹In-J591 SPECT to detect isolated intraprostatic foci, likely secondary to inherent SPECT resolution limits. Clinically significant disease (Gleason score ≥7) was readily detected by immuno-PET, and may potentially serve as a non-invasive means for active surveillance monitoring. Based on the success of these preliminary studies, we are initiating a phase 2 immuno-PET/MRI study to further evaluate the multitude of uses of one of the first successful molecular imaging agents in localized prostate cancer.