In men with high-risk prostate cancer, imaging with prostate-specific membrane antigen (PSMA) PET-CT prior to curative surgery or radiotherapy proved far more accurate than conventional imaging for detecting metastatic disease, a randomized trial found.
Among 295 evaluable patients, gallium-68 PSMA-11 PET-CT imaging had an accuracy of 92% (area under the curve [AUC] of the receiver operating characteristic curve) versus 65% with standard CT imaging and bone scans (P<0.0001), reported Michael Hofman, MBBS, of the Peter MacCallum Cancer Centre in Melbourne, Australia, and colleagues in The Lancet.
There were fewer equivocal cases of metastatic disease with PSMA PET/CT (7% vs 23% with conventional imaging), and the novel imaging method altered the course of disease management for twice as many men following first-line imaging (28% vs 15%, P=0.008). In those who underwent second-line imaging — for men with no more than two unequivocal metastases — change in treatment occurred in 27% of those that crossed over to PSMA PET/CT, but just 5% of those switching to conventional imaging.
“Taken together, our findings indicate that PSMA-PET/CT scans offer greater accuracy than conventional imaging and can better inform treatment decisions,” Hofman said in a statement. “We recommend that clinical guidelines should be updated to include PSMA PET/CT as part of the diagnostic pathway for men with high-risk prostate cancer.”
PSMA “is a cell-surface glycoprotein overexpressed on prostate cancer cells,” according to background information in the paper. Using radiolabelled small molecules that bind to these cells, PSMA PET/CT can detect tumor presence throughout the body.
Patients receiving PSMA PET/CT were also exposed to significantly lower levels of radiation than with conventional imaging (8.4 vs 19.2 millisieverts, respectively).
“Current medical imaging techniques often fail to detect when the cancer has spread, which means some men are not given the additional treatments they need,” co-author Declan Murphy, MBBCh, also of Peter MacCallum, said in a statement. “Our findings suggest PSMA-PET/CT could help identify these men sooner, so they get the most appropriate care.”
In an accompanying editorial, Caroline Moore, MD, of University College London, pointed to one limitation in the study — the way metastasis was defined — that may have biased the results in favor of PSMA PET/CT.
Metastatic disease was defined as either a bone lesion changing to sclerotic or blastic on follow-up imaging, or histologic confirmation of a metastatic site — this “hard criteria” was met in just 23% of the 87 men in the study with metastases. Alternatively, metastasis could also be established via three “soft criteria,” which included rising prostate-specific antigen (PSA) levels at 6 months, increases or decreases in lesion number or size on subsequent imaging, a lesion associated with clinical symptoms, and others.
“Although these criteria reflect a real-world approach, in which men having radiotherapy will not have histological confirmation of nodal disease, some men might have had microscopic disease that was not detected by either modality,” wrote Moore. “Also, some men might have had false-positive findings, which were never further assessed.”
She noted that in the study, a small group of men still went on to radical treatment despite the presence of metastatic disease on PSMA PET/CT.
“Introduction of new imaging modalities, such as PSMA PET-CT, with improved sensitivity in detecting small-volume metastases, brings both challenges and opportunities,” she continued. “In particular, when men test negative on conventional imaging, and PSMA PET-CT shows small-volume metastatic disease, what should we do?”
From 2017 to 2019, the proPSMA study randomized 302 men with high-risk prostate cancer 1:1 to either standard CT imaging plus a bone scan or to PSMA PET-CT across 10 sites in Australia. Median participant age was 68. High-risk disease was defined as one of the following: clinical stage ≥T3, a PSA level ≥20 ng/mL in the 12 weeks prior to randomization, or International Society of Uropathology grade group 3-5.
Of the 295 men with follow-up, 30% had either pelvic nodal metastases or distant metastases. Sensitivity was improved with PSMA PET/CT, at 85% compared to 38% with conventional imaging. As was specificity (91% vs 98%, respectively).
In subgroup analyses, PSMA PET/CT was more accurate at detecting both pelvic nodal metastases (91% vs 59% with conventional imaging) and distant metastases (95% vs 74%).
The study was funded by Movember and the Prostate Cancer Foundation of Australia.
Hofman disclosed grants from the Prostate Cancer Foundation of Australia, Movember, the Peter MacCallum Foundation, the U.S. Department of Defense, and the Victorian Cancer Agency; and other relationships with Ipsen, Sanofi Genzyme, and Janssen. Coauthors reported grants from or financial relationships with the Prostate Cancer Foundation of Australia, Mundipharma, Janssen, Ferring, Telix Pharmaceuticals, Astellas, Janssen, Bayer, and Ipsen.
Moore reported financial or advisory board relationships with Janssen, Astellas, Sonablate, Steba Biotech, and Genomic Health.