Should an MRI Come Before the Prostate Biopsy? Reading the PRECISION Trial

The short answer#

PRECISION was a 500-man randomized trial, published in the New England Journal of Medicine in 2018, that asked a simple sequencing question: when a man has a suspicious PSA and has never been biopsied, does scanning the prostate first change the diagnosis compared with going straight to biopsy? Putting a multiparametric MRI at the front of the pathway found more of the cancers worth treating, found fewer of the harmless ones, and let a substantial share of men avoid a biopsy entirely. That is what the trial measured in the men it enrolled. It is not a recommendation for you.

Key points#

• PRECISION compared two pathways after a suspicious PSA: MRI first, biopsy only if the scan flagged something, versus standard ultrasound-guided biopsy for everyone.
• The MRI-first pathway diagnosed clinically significant cancer in 38 percent of men, against 26 percent with the standard route.
• It also diagnosed fewer insignificant cancers (9 percent versus 22 percent), reducing overdiagnosis.
• About 28 percent of the MRI group had a reassuring scan and skipped biopsy altogether.
• The gains depend on high-quality scans and skilled readers, and the trial did not track long-term survival.

The problem PRECISION was built to solve#

Start with what a biopsy was actually doing before MRI entered the picture. The old default after a worrying PSA or an abnormal exam was a transrectal ultrasound-guided biopsy, a systematic sampling of the gland with roughly 10 to 12 needle cores. The catch is that ultrasound cannot see most prostate tumors. The needles follow the shape of the gland, not the location of disease, so the procedure is closer to sampling a field in a grid than aiming at a target.

That blind grid fails in two opposite directions at once. It can walk right past an aggressive tumor sitting in a corner the needles happened to miss, producing a falsely reassuring result. And it can turn up a small, indolent cancer that would never have caused symptoms in a man's lifetime, pulling him toward surgery or radiation, and their lasting side effects, for a disease that did not need to be found. Missing the dangerous cancers and over-finding the trivial ones are both real costs.

Multiparametric MRI raised an obvious alternative. If a scan could highlight suspicious tissue before any needle went in, two things might follow: biopsies could be aimed at real targets, and men whose scans looked clean might reasonably skip the needle. PRECISION set out to test that idea against the old pathway directly, rather than assume it.

How the trial was set up#

The investigators, reported by Kasivisvanathan and colleagues and registered as NCT02380027, enrolled 500 men at 23 centers across 11 countries in Europe and North America. Every man had a clinical suspicion of prostate cancer, from an elevated PSA or an abnormal exam, and none had been biopsied before. Each was randomly assigned to one of two routes.

In the standard route, a man went directly to ultrasound-guided systematic biopsy, the familiar 10 to 12 cores. In the MRI route, he had a multiparametric MRI first. If the scan showed a suspicious area, he had a biopsy aimed only at that area, with no systematic sampling added. If the scan showed nothing suspicious, he had no biopsy at all and was followed.

The main question was the share of men found to have clinically significant cancer, which the trial defined as a Gleason score of 3+4 (a total of 7) or higher. That threshold is the whole point. Nobody doubts MRI can find more cancer of some kind. The design deliberately asked whether it finds more of the cancer that actually warrants acting on.

The three findings that made it matter#

First, detection of the cancers that count went up. Clinically significant cancer was diagnosed in 38 percent of the MRI group versus 26 percent of the standard group. After adjustment, that came to roughly a 12 percentage point advantage for the MRI-first pathway, and it was statistically significant.

Second, and just as important, overdiagnosis went down. The MRI route diagnosed clinically insignificant cancer, the low-grade disease most likely to generate worry and treatment without benefit, in 9 percent of men, against 22 percent with standard biopsy. So the pathway did not simply find more; it shifted the mix toward disease that matters and away from disease that mostly harms through the act of finding it.

Third, many men avoided the procedure. Of the men assigned to the MRI route, 71 of 252 (about 28 percent) had scans that were not suspicious and therefore had no biopsy. A prostate biopsy is not a small thing; it carries risks of infection, bleeding, and discomfort. Sparing more than a quarter of men that step while still catching more significant cancer is the pairing that made the trial influential.

Where the trial stops#

A clean randomized design still measures only what it was built to measure, under the conditions it was run in, and those boundaries deserve as much attention as the headline. PRECISION enrolled men with no prior biopsy, so the results do not automatically carry over to men being re-evaluated after an earlier negative biopsy, or to men already on active surveillance. The MRI route also leaned on high-quality scans read by experienced radiologists at academic centers. In routine practice, scan quality and reader skill vary, and a pathway where a negative scan cancels the biopsy is only as trustworthy as the scan behind it.

There is a genuine trade buried in that 28 percent who skipped biopsy. If you decline the needle after a reassuring MRI, you are accepting a small chance that a significant cancer was present but not seen. PRECISION was not designed or sized to follow long-term outcomes such as spread or survival; it counted what was detected at the point of diagnosis. Later reviews and follow-on analyses have tried to estimate how many significant cancers an MRI-first strategy might miss, and the fair summary is that the number is low but not zero. That balance is why guideline bodies, including NICE in England, moved toward recommending multiparametric MRI before biopsy for suspected localized prostate cancer while keeping systematic safeguards in place.

None of this hands you an answer. Whether an MRI belongs before a biopsy in a particular case turns on the reason for suspicion, the quality of local imaging, prior test history, and how you weigh the risk of a missed cancer against the risk of an unnecessary procedure. Those judgments belong to you and your own clinician. What PRECISION established is narrower and still worth carrying: in men not biopsied before, a good scan at the front of the pathway changed the mix of what got diagnosed, toward significant cancer and away from insignificant cancer, and let many men skip the needle without an obvious cost in detection.