A negative MRI is a probability, not a verdict, and a new review asks whether PSMA PET can close that gap or whether it is just another mirage.

The Clinical Problem

I wear two hats in my clinical work. The first is helping men navigate the prostate cancer puzzle. That means giving them clarity about what they actually have, which is often different from what they think they have. It also means helping them decide what comes next: biopsy, treatment, or surveillance.

The second is giving them the best natural and lifestyle strategies to optimize their health, reduce their risk of dying from the disease, and ideally live better than they did before the diagnosis.

Both hats depend on the same thing: an accurate picture.

I’m writing this from the Focal+ meeting, surrounded by an exceptional group of prostate cancer experts. I’m preparing to give my talk, Beyond Treatment: Lifestyle Choices That Impact Prostate Cancer, and it’s a privilege to be here.

One presentation stayed with me. It made a point every focal therapy clinician already feels: Treatment is only as good as the images.

You know the patient. His PSA keeps rising, or his PSA density is high, or his father died of prostate cancer. His MRI comes back negative or equivocal, PI-RADS 2 or 3, and your gut says it isn’t the whole story. Do you biopsy anyway? Repeat the MRI? Watch and wait?

That is the gap this issue addresses.

Why This Matters — To You and Your Patient

MRI changed prostate cancer diagnosis. It is not perfect. A meta-analysis cited in the review reported an MRI negative predictive value (NPV) of 88%. That means about 88 of every 100 men with a negative MRI did not have clinically significant prostate cancer (csPCa). That sounds reassuring until you look at who is being scanned.

When cancer prevalence reached 60%, NPV fell to 67%. In a higher-risk man, a negative MRI means much less. NICE guidance cited by the authors estimates that 11% to 28% of men with a “negative” MRI may still have csPCa.

Read that again. Up to roughly 1 in 4.

For the patient, this cuts both ways. A missed cancer means false reassurance. An unnecessary biopsy means pain, infection risk, anxiety, and cost.

For focal therapy, the stakes are higher still. If we treat the lesion we can see and miss the one we can’t, the treatment was built on an incomplete map.

How It Works — Mechanism & Physiology

MRI and PSMA PET look for cancer in fundamentally different ways.

MRI reads architecture. It detects the structural consequences of a tumor: increased cell density, new blood vessels, and remodeling of surrounding tissue. To show up on MRI, a tumor has to become large and biologically developed enough to change the landscape.

PSMA PET reads a molecular signal. Prostate-specific membrane antigen is a protein on the cell surface that is often overexpressed in prostate cancer. Its expression tends to rise with Gleason score and tumor aggressiveness.

We attach a radioactive tracer, such as 68Ga or 18F, to a molecule that binds PSMA. The tracer emits positrons, and the PET scanner detects where the tracer has gathered.

Here is the analogy I find useful.

MRI is like satellite imagery of a city. It sees buildings once they are built. PSMA PET is like detecting cell phone signals. It can tell you people are gathering before the building goes up.

This leads to the authors’ central hypothesis: some PSMA-positive, MRI-negative cancers may be sending a molecular signal before they cause enough structural change to appear on MRI.

Two important caveats.

First, this is a hypothesis, not a proven natural history. 

Second, there is a simpler explanation. Some “MRI-occult” tumors may reflect suboptimal MRI quality or reader experience. They might have been visible at an expert center.

Preliminary imaging and protein-expression research suggests the two scans may measure partly different biology. PSMA-derived signals have been associated with tumor heterogeneity and metabolic activity. The authors stress that this work is early.

The Therapeutic Goal

The goal is not simply to “find more cancer.”

We already find plenty of cancer that doesn’t need treatment. The goal is to find the right cancer, and to confidently avoid biopsying the wrong patient.

That breaks into two targets:

Confident omission. Can a negative combined assessment be reassuring enough to skip or delay biopsy safely?

Accurate localization. When cancer is present, can we map it well enough to target the biopsy, and for focal therapy, the treatment?

For my second hat, there is a third target. A man who knows what he actually has can make grounded decisions about treatment and about how he lives. Clarity is therapeutic in itself.

What the Evidence Says

The review by Wagner, Emberton, Kasivisvanathan, Punwani, and colleagues pulls together the prospective data on PSMA PET in men with negative or equivocal MRI. The title asks the honest question: is PSMA PET the answer to MRI-occult prostate cancer, or another mirage?

Brief Aside: Dr. Mark Emberton and Dr. Veeru Kasivisvanathan have both been guests on my podcast. I have no financial relationship with the authors or with any PSMA tracer manufacturer. 

Go deeper with the authors:

Skip MRI contrast Gadolinium, Precision Biopsy, UK vs. US Care, Dr. Veeru Kasivisvanathan — Episode 179

Why MRI & PSMA PET scans are better with Dr. Mark Emberton — Episode 170

Here is what the individual studies found.

PRIMARY trial (the strongest signal). Among men with PI-RADS 2 MRI, 28% (27/95) had csPCa. Among PI-RADS 3, 47% (25/53) had csPCa. PSMA PET/CT was positive in 90% (47/52) of those cancers, or about 9 of every 10.

In the broader comparison, adding PSMA PET/CT to MRI raised sensitivity from 83% to 97%. NPV rose from 72% to 91%. Positive predictive value (PPV) was preserved.

In plain terms: the combined approach caught about 97 of every 100 men with csPCa, versus 83 with MRI alone.

Shi et al. In 56 men with PI-RADS 3 lesions, 14% had csPCa. PSMA PET/MRI sensitivity was 87.5%, specificity 83.3%, and NPV 97.5%. But PPV was only 46.7%. Fewer than half of positive scans were real csPCa.

PEDAL study (the counterweight). Among men with PI-RADS 1–2 MRI, 18% (11/61) had csPCa. PSMA PET/CT found 82% of them, roughly 8 in 10.

But in PI-RADS 3, PSMA PET/CT caught only 2 of 6 cancers (33%). Across PI-RADS 1–3, it detected 65% of csPCa. The authors report that MRI actually had higher diagnostic accuracy than PSMA PET/CT in PEDAL, though the difference was not considered clinically significant.

Ferraro et al. PSMA PET/MRI showed 96% sensitivity, 81% specificity, 89% PPV, 93% NPV, and 90% accuracy. Impressive, but only 7 patients had PI-RADS 3 lesions, and only 1 had csPCa.

Lopci et al. Among 15 men with negative MRI, PSMA-guided biopsy found csPCa in 4 (26.7%), roughly 1 in 4.

Privé et al. Among 26 men with PI-RADS 3 lesions, PSMA PET/CT upgraded 12% and downgraded 54%. NPV was 93%, but PPV was only 27%, about 1 in 4 positive scans.

The pattern: sensitivity and NPV are generally high. The size of the benefit varies a lot between studies. The authors’ conclusion is that PSMA PET looks complementary to MRI, not a replacement for it.

The evidence gaps, named plainly: these are mostly small cohorts, they use different positivity thresholds, and they report diagnostic accuracy only. None tell us whether finding these cancers changes survival, metastasis, or quality of life.

What It Does NOT Do

It does not replace MRI. No study here supports skipping MRI in favor of PSMA PET.

A positive scan is not a diagnosis. In equivocal lesions, PPV ranged from 27% to 47%. Most positive scans in some cohorts were not csPCa. A positive PSMA result earns a targeted biopsy, not a treatment plan.

It is not standardized yet. PRIMARY 2 calls PRIMARY scores 3–5 positive. Shi counted only scores 4–5. PEDAL used SUVmax cutoffs plus a reader-certainty scale. The same scan could be called positive at one center and negative at another.

It is not consistently superior. PEDAL is the reminder. In PI-RADS 3 patients, it missed 4 of 6 cancers.

It does not prove the “molecular before structural” story. That remains a hypothesis. Some MRI-occult tumors may simply reflect a suboptimal MRI.

It does not yet have outcome, cost, or cost-effectiveness data. We don’t know whether routine use improves what matters most to patients.

Safety, Practical Use and Clinical Considerations

This is a diagnostic imaging question, not a treatment one, so there are no drug doses or therapeutic exposures to discuss. The practical questions are about who, when, and how to interpret.

Who to consider. The man with persistent suspicion despite a negative or equivocal MRI. The review notes clinicians already use PSA density, PSA velocity, and family history to decide when concern remains after a negative MRI. NICE examples include PSA density above 0.15 ng/mL/mL and PSA velocity above 0.75 ng/mL/year.

Question the MRI first. Before ordering a second test, ask whether the first one was good enough. An expert re-read or a repeat MRI at a high-volume center may resolve the question.

Know your center’s scoring system. Ask whether your reading center uses the PRIMARY score, SUVmax thresholds, or another suspicion scale, and where it draws the line for “positive.”

Treat a positive scan as a targeting tool. In PRIMARY 2, positive patients receive targeted transperineal biopsy of up to 4 PSMA-positive lesions.

Treat a negative combined assessment as reassurance, not a discharge. PRIMARY 2 follows PSMA-negative patients for 2 years without immediate biopsy. In Privé et al., men with PI-RADS 1–2 and low PSMA suspicion were followed with PSA for 12 months, with repeat MRI if PSA rose by more than 25%.

Individualize. The content reviewed here does not address radiation exposure, tracer-specific differences, access, or cost. Those conversations belong with each patient and your imaging team.

Final Thoughts

Every good decision in prostate cancer starts with an honest map. 

PSMA PET may add a molecular layer that MRI can’t see, but a new layer of information is not the same as a better outcome, and the evidence hasn’t earned that claim yet. 

Evidence first, context always: our job is to use every tool that brings clarity, while staying humble about what each one can and cannot tell us.

Actionable Clinical Takeaways

1. Treat a negative MRI as a probability shaped by the patient’s risk. In higher-risk men, NPV can fall to about 67%, so factor in PSA density, velocity, and family history before you reassure. 
2. Audit MRI quality before adding PSMA PET. An expert re-read may answer the question more simply and cheaply.
3. Use PSMA PET as a complement, never a substitute. The strongest data, from PRIMARY, show benefit when it is combined with MRI, with sensitivity rising from 83% to 97%.
4. Biopsy a positive PSMA scan before acting on it. In equivocal lesions, as few as 1 in 4 positive scans represented csPCa.
5. Follow a negative combined workup with structured surveillance. Use PSA monitoring and a clear trigger for repeat imaging, such as a PSA rise above 25%, rather than a one-time all-clear.

Reference

Wagner, T, Emberton, M, Caglic, I, Priftakis, D, Giganti, F, Kasivisvanathan, V, & Punwani, S (2026). PSMA PET: The answer to MRI-occult prostate cancer… or another mirage? European Radiology, 36(9), 7063–7072. DOI: 10.1007/s00330-026-12596-3