PRECISION data lend validation to utility of multiparametric MRI

May 29, 2018

"This important study provides strong validation of the utility of multiparametric MRI of the prostate prior to initial biopsy," writes Badar M. Mian, MD.

“Journal Article of the Month” is a new Urology Times section in which Badar M. Mian, MD (left), offers perspective on noteworthy research in the peer-reviewed literature.  Dr. Mian is associate professor of surgery in the division of urology at Albany Medical College, Albany, NY.

In men with clinical suspicion of prostate cancer (elevated PSA or abnormal DRE), multiparametric magnetic resonance imaging of the prostate may be useful as a triage test to reduce the number of unnecessary biopsies while enhancing the detection of clinically significant prostate cancer (CS-PCa). The recent paper from the PRECISION study group reported that, in biopsy-naïve men, CS-PCa detection (any core with Gleason score ≥3+4) was significantly higher and the risk of cancer detection was significantly lower in men undergoing MRI-targeted biopsy when compared to the standard transrectal ultrasound-guided biopsy (N Engl J Med March 18, 2018 [Epub ahead of print]).

The authors identified 500 patients with elevated PSA and/or abnormal DRE and randomized 252 men to the MRI-targeted biopsy group and 248 to the standard-biopsy group. Of the 252 men in the MRI-targeted biopsy group, 181 (72%) with suspicious lesions (defined as PIRADS 3-5) underwent biopsy of the target lesions only while 71 men (28%) with a normal MRI (PIRADS 1 or 2) did not undergo a biopsy at all. While the standard-biopsy group had systematic biopsy with an average of 12 cores, the MRI targeted biopsy obtained an average of four cores per patient.

Ninety-five men (38%) in the MRI-targeted biopsy group had CS-PCa compared to 64 (26%) in the standard-biopsy group (p=.005). Fewer men were diagnosed with low-risk prostate cancer in the MRI-targeted biopsy group (23 men, 9%) than in the standard-biopsy group (55 men, 22%, p<.001).

In men with the MRI demonstrating a targetable lesion, 51 of 175 (29%) had a PI-RADS score 3, 70 (40%) had a score of 4, and 54 (31%) had a score of 5. CS-PCa rate for PIRADS score 3, 4, and 5 was 12%, 60%, and 83%, respectively. Conversely, the negative biopsy rate for PIRADS score 3, 4, and 5 was 67%, 31%, and 6%, respectively.

Percentage of biopsy cores with cancer was significantly higher in the MRI-targeted biopsy group (422 of 967 cores, 44%) than in the standard-biopsy group (515 of 2,788 cores, 18%); ie, much fewer cores were needed to detect a higher rate of CS-PCa. Post-biopsy complications were less common in the MRI-targeted group than the standard-biopsy group, including hematuria (30% vs. 63%), hematospermia (32% vs. 60%), pain (13% vs. 23%), rectal bleeding (14% vs. 22%), and erectile dysfunction (11% vs. 16%), respectively. This difference in the patient-reported 30-day complication rate is explained by the fact that fewer men in the MRI-targeted group needed a biopsy and only a few cores were obtained per biopsy session.

Next:Better results for primary, secondary outcomesBetter results for primary, secondary outcomes

This randomized controlled trial was designed to demonstrate “non-inferiority” of the MRI-targeted approach when compared to the standard biopsy approach. But it appears that the MRI-targeted approach was superior and yielded better results for primary and secondary outcomes, especially the detection of CS-PCa in biopsy-naïve men.

In order to allow for future applicability of the results to the wider population, the trial was designed to include patients and physicians from 25 sites in 11 countries. Only 14% of participants were from North America and 86% were from European sites, reflecting the higher availability and utilization of MRI of the prostate in Europe. Over 50% of the study sites enrolled fewer than 10 participants per site. There was significantly heterogeneity in the biopsy technique, which was left at the discretion of each site, based on their local expertise and availability. These differences included MRI (1.5T vs. 3.0T scanner), use of endorectal coil, fusion technique (software vs. visual), use of five different MRI-US fusion platforms, and transrectal versus transperineal biopsy.

Since most of the participants and physicians were from higher volume centers, with previously demonstrated expertise in MRI-targeted biopsies, it is not clear whether these results can be generalized to a non-trial setting. It is notable that despite having reviewed and agreed to the protocol for performing and reporting the MRI findings, a concordant score between the local and central review was noted in only 78% of the MRIs.

This important study provides strong validation of the utility of multiparametric MRI of the prostate prior to initial biopsy. This report does not address the CS-PCa that may be missed in men with non-suspicious MRI and no biopsy. Important research often yields opportunities to address clinically relevant questions. Whether the 12% increased detection of CS-PCa using MRI-targeting is weighted favorably against the 8%-10% missed CS-PCa (other reports) that are missed will be subject of future studies.