The accuracy was 92% for PSMA PET/CT and 65% for CT and bone scintigraphy (P < .001), according to data reported at the virtual annual congress of the European Association of Urology and published in The Lancet.
In addition, PSMA PET/CT had greater effects on treatment. First-line imaging led to treatment changes in 28% of the PSMA PET/CT group and 15% of the CT/bone scan group. Second-line imaging led to treatment changes in 27% and 5% of patients, respectively.
“My strong view is that this is practice-changing data,” said study investigator Michael Hofman, MBBS, of the Peter MacCallum Cancer Centre in Melbourne.
Highly relevant secondary outcomes were included in the study, Dr. Hofman said, and results were all in favor of PSMA PET/CT over conventional imaging.
PSMA PET/CT was associated with a lower rate of equivocal or uncertain findings (7% vs. 23%), and half the radiation dose was needed with PSMA PET/CT (8 mSv vs. 19 mSv). Furthermore, PSMA PET/CT was more accurate when used after CT/bone scan than when CT/bone scan was used after PSMA PET/CT (19% vs. 2%).
“PSMA PET/CT has emerged as a potential new gold standard for imaging prostate cancer,” Dr. Hofman said. The images it can produce were “striking” compared to conventional CT, he added. Pelvic and abdominal metastases that are barely visible on CT were “lighting up very brightly” on PSMA PET/CT, he said.
The study also showed that PSMA PET/CT was superior to CT/bone scans for picking up metastases throughout the body. The detection rate was 91% and 59%, respectively, for pelvic nodal metastases and 95% and 74%, respectively, for distant metastases.
Study details
ProPSMA is a multicenter, phase 3 trial directly comparing PSMA PET/CT and the standard of imaging. Of 339 men assessed for inclusion across 10 centers in Australia, 302 were randomized. They had a median age of 69 years. All patients had high-risk prostate cancer, which was defined as a prostate-specific antigen level of 20 ng/mL, Gleason Grade Group 3-5, or clinical stage T3 or higher. They were all about to undergo either surgery or radiotherapy with the intention of curing their prostate cancer.
PSMA PET/CT was performed using the gallium-68-labelled PSMA-11 tracer, but the results would likely be no different if another tracer were used, Dr. Hofman said in the discussion following his talk.
Of the three available tracers, there were minor differences, mostly in how they were excreted. However, “they’re all extremely good. I’m not sure anyone’s ever going to undertake a head-to-head study comparing them,” Dr. Hofman said.
“Whichever one you can access, at the cheapest cost, I think, is going to be the best one in your center,” he added. “That really does vary geographically, but I really don’t think one is better or worse than the other.”
Praise and criticism
The latest European guidelines acknowledge that PSMA PET/CT is more sensitive for detecting lymph node and bone metastases than the classical workup of abdominopelvic CT and bone scintigraphy, according to invited discussant Matthias Heck, PD Dr. med, of the Technical University of Munich in Germany.
“Molecular imaging using PSMA PET/CT facilitates the detection of small lymph node metastasis, with the size of a few millimeters,” Dr. Heck said.
Although he commended the ProPSMA investigators, Dr. Heck had one criticism of the study design that may have resulted in over-sensitivity of PSMA PET/CT.
“As a urologist, I want to address as a discussion point the low number of histopathologic validation in the ProPSMA study,” he said. “Pelvic lymph node sampling was performed only in 66% of patients treated with radical prostatectomy for high-risk prostate cancer. Hard criteria to define the presence of metastasis were only used in 23% of patients with metastases. Therefore, it is possible that the sensitivity was overestimated by using mainly soft criteria.”
The sensitivity of PSMA PET/CT was 85%, while that of CT/bone scan was 38%. The respective specificities were 98% and 91%.
“What I like most about this study is that, when we perform a PSMA PET/CT, you see the whole body; you don’t see only pelvic lymph nodes,” Dr. Heck said. Since it was not possible to validate distant metastasis by histopathology, he added, this imaging method could clearly help determine the best treatment.
“If we have distant metastasis in the bones or in the lymph nodes outside of the pelvis, it’s clearly unnecessary to direct this patient to undergo local treatment, and we need to think about other treatments,” Dr. Heck said. “Therefore, I think it’s a very important question that is being raised by this study, and we all need to look at the whole body of the patient and not focus only on the pelvic lymph nodes.”
The study was funded by the Prostate Cancer Foundation of Australia. Dr. Hofman said he has no relevant conflicts of interest. Dr. Heck disclosed relationships with Astellas, Janssen, Ipsen, Amgen, Bayer, Heise, Merck, Sanofi, and Takeda.
SOURCES: Hofman M et al. Lancet. March 22, doi: https://doi.org/10.1016/S0140-6736(20)30314-7.