PSMA PET for Prostate Cancer: How the proPSMA Trial Changed Staging

The proPSMA trial, a prospective randomised study run across 10 Australian centres and published in The Lancet in 2020, showed that a single PSMA PET-CT scan stages high-risk prostate cancer far more accurately than the long-standard pairing of CT and bone scan. Reading the scans of 302 men, the investigators found PSMA PET-CT reached 92% accuracy for detecting spread to lymph nodes and distant sites, against 65% for conventional imaging, a 27 percentage-point gap, while exposing patients to less than half the radiation. That single result is a large part of why PSMA PET moved from research novelty to a front-line staging tool, and why the U.S. Food and Drug Administration approved the first PSMA imaging agent later the same year. The trial did not prove that acting on those extra findings lengthens life, and it enrolled only men with high-risk disease.

What proPSMA actually tested

For decades, staging newly diagnosed high-risk prostate cancer meant a CT scan of the abdomen and pelvis plus a technetium bone scan. Both are indirect: CT infers nodal disease from lymph-node size, and bone scintigraphy detects the bone's reaction to a tumour rather than the tumour itself. PSMA imaging works on a different principle. Prostate-specific membrane antigen is a protein expressed at high density on most prostate cancer cells, and a radiolabelled molecule that binds it lights up cancer deposits directly on PET, even small ones.

proPSMA, led by Michael Hofman and colleagues, randomised men with high-risk disease who were candidates for curative surgery or radiotherapy to first-line imaging with either conventional scans or gallium-68 PSMA-11 PET-CT. The design had a useful feature: after the first scan, men crossed over to the other arm unless three or more unequivocal distant metastases were already found. That crossover let the same patients serve in both imaging pathways and gave the trial a fair comparison against a composite reference standard built from pathology, follow-up imaging, and clinical course.

The numbers, and why the design matters

The primary endpoint was accuracy for detecting pelvic nodal and distant metastatic disease, expressed as area under the curve. PSMA PET-CT scored 92% (95% CI 88 to 95) versus 65% (95% CI 60 to 69) for CT plus bone scan. The gain came from both directions: sensitivity rose from 38% to 85%, and specificity from 91% to 98%. In plain terms, conventional imaging missed real disease and was more often equivocal, with 23% of studies returning uncertain findings compared with 7% for PSMA PET-CT.

Two secondary findings carry practical weight. First, PSMA PET-CT changed planned management more often, in 28% of men versus 15% with first-line conventional imaging. Finding disease the older scans could not see redirected patients away from operations or radiation fields that would not have cured them. Second, radiation exposure was substantially lower: 8.4 millisieverts for the PSMA pathway versus 19.2 for conventional imaging, less than half the dose. A more accurate test that also halves radiation is an uncommon combination, and it is the reason the trial landed as firmly as it did.

The regulatory response followed. On December 1, 2020, the FDA approved gallium-68 PSMA-11, the first PSMA-targeted PET agent, for two uses: initial staging of men with suspected metastasis who are candidates for curative therapy, and evaluation of suspected recurrence signalled by a rising PSA. That approval, documented in the peer-reviewed literature, broadened access in the United States.

What the trial did not settle

Strong evidence still has edges, and reading them honestly is the point of appraisal. proPSMA measured diagnostic accuracy and short-term management change. It did not, in its primary report, demonstrate that patients live longer or relapse less because their care was guided by PSMA findings. Detecting more disease reliably reshuffles who is labelled metastatic, a form of stage migration in which apparent outcomes can improve for each group without anyone actually being cured. The profession's appropriate-use criteria flag exactly this effect. A better map is valuable, but it is not the same as a better destination.

The biology sets limits too. PSMA expression is high in most prostate cancers, yet a minority of tumours express little of it, and poorly differentiated or neuroendocrine disease can be PSMA-negative and slip past the scan. In the other direction, PSMA uptake is not unique to prostate cancer: normal ganglia, healing rib fractures, degenerative bone, and some other tumours can take up the tracer and generate false positives. This is why standardised reading criteria and experienced interpretation matter, and why a single suspicious focus usually warrants correlation rather than an immediate change of plan.

Scope matters as well. proPSMA enrolled high-risk men, so its results speak to that group. They are not a warrant for scanning low-risk or favourable-intermediate-risk cancers, where the pre-test probability of metastasis is low and incidental uptake is more likely to mislead than to help. The appropriate-use question is whether the added information will actually change a decision for the patient in front of you. Greater sensitivity alone does not answer that.

Where PSMA PET fits now

The durable lesson of proPSMA is that a direct molecular test outperformed two indirect anatomical ones for staging high-risk disease, with fewer ambiguous results and lower radiation. Major guidelines and formal appropriate-use criteria have since incorporated PSMA PET into initial staging and biochemical-recurrence workups, and its role in selecting patients for PSMA-directed radioligand therapy has extended the same targeting principle from diagnosis into treatment. The technology earns its place when it answers a question that changes management, and it earns skepticism when it is used reflexively in settings the evidence never tested. This article is educational and not medical advice; staging decisions belong to a patient and their own clinical team.