First Trial of Targeted MRI vs Standard Prostate Biopsy

Kate Johnson

April 18, 2014

STOCKHOLM — The first prospective trial comparing traditional transrectal ultrasound (TRUS)-guided biopsy with targeted MRI-guided biopsy for the triage of patients with suspected prostate cancer lands soundly in favor of MRI, Australian researchers reported here at the European Association of Urology (EAU) 29th Annual Congress.

"MRI has now become the investigation of choice for all men with suspected prostate cancer; it has completely radicalized and changed our paradigm," said Morgan Pokorny, MD, a urologist from the Wesley Hospital in Brisbane, Australia.

The results of this study, also published online March 14 in European Urology, generated a flurry of reaction from an audience consisting primarily of urologists, who are perhaps uncomfortable with sharing their specialty with radiology.

"One of the key messages is that urologists and radiologists need to become best friends," Dr. Pokorny told the gathering.

"I think the future lies with MRI," he said later to Medscape Medical News. "We're not categorically saying you shouldn't do TRUS-guided biopsies, but we think that MRI can help you risk-stratify very powerfully — more so than any other tool."

Although few urologists would argue with that general idea, there is a flip side, said Boris Hadaschik, MD, from the University of Heidelberg in Germany, who chaired the session during which the results were presented.

"Everybody appreciates the great potential of MRI to help us see significant lesions. MRI is the best tool we have to find clinically relevant cancer. However, it always depends on what your priorities are," he told Medscape Medical News.

"If you really want to reduce overdetection, if you do not want to find the indolent tumors and don't object to missing 5% to 10% of the significant tumors in the first round of biopsies, then targeted MRI biopsies are the way to go initially," said Dr. Hadaschik.

However, when it is done well, traditional systematic TRUS-guided biopsy can rival targeted MRI procedures for prostate cancer detection, according to another study, presented by Andreas Hiester, MD, from the University of Dusseldorf in Germany.

"Both biopsy methods showed an equally high prostate cancer detection rate of 53.1% and a similar detection rate of clinically significant cancer," he said.

But that study was criticized for having a "bias toward an improved control arm," and Dr. Hadaschik said he wished the 2 studies had been published alongside each other.

Australians Show Superiority

The Australian study that showed the superiority of targeted MRI-guided biopsy was a single-center prospective study conducted in 233 consecutive biopsy-naïve subjects with elevated levels of prostate specific antigen (PSA). All the participants underwent multiparametric (mp)MRI, and the images were then evaluated using the Prostate Imaging Report and Data System (PI-RADS).

After this evaluation, subjects who were not suspected of having significant prostate cancer received a standard 12-core TRUS-guided biopsy. Subjects whose MRI was suggestive of significant cancer underwent targeted MRI-guided biopsy with 2 to 4 core samples, followed by the standard TRUS-guided biopsy performed by a urologist who was blinded to the MRI findings.

For patients with intermediate- to high-risk cancer, the estimated sensitivity of MRI plus biopsy was 92.34% — "far outperforming that of the TRUS plus biopsy (70.44%)," said Dr. Pokorny. Specificity was high in both cases — 96% and 93%, respectively – and the relative risk of detecting intermediate- to high-risk cancer (PI-RADS 4 to 5) was 14.4.

"The design of our study allowed us to look back and see what would have happened if we had only biopsied men with suspicious lesions (PI-RADS 4 to 5) and didn't do any random biopsies on men with normal scans," he said. "If we adopted this approach, we could reduce the number of men needing biopsies by 50% (223 vs 109), we could reduce the biopsy cores we took by 80% (2672 vs 322) and, most importantly, we could reduce the diagnosis of low-risk prostate cancer by up to 90% (47 vs 5), and still we'd find 12% more intermediate- to high-risk cancers (79 vs 89)."

"Overall, our study showed that you find more significant cancers with MRI than with TRUS-guided biopsy," Dr. Pokorny told Medscape Medical News. "The 2 together will give you the best outcome, but I think as a society and as professionals, we should move toward a goal of trying to do less on men than we're currently doing. In other words, biopsy fewer men and put fewer needles into prostates."

"The only question that we can't answer right now is whether we should omit the systematic, conventional 12-core nontargeted biopsies," said Dr. Hadaschik. "If you omit them, you will miss a few significant tumors, but you will greatly reduce overdiagnosis of indolent tumors. You have to counsel the patient by saying, 'If you want maximum security, then you need both systematic and targeted biopsies or, if you are okay with us only looking for significant tumors, then we will continue to watch and maybe find a significant tumor in a year or 2. Nobody knows the clinical relevance if you wait a little bit longer'."

The Australian study was very unusual, in that the MRI biopsies were "in-bore," meaning they were performed while the patient was in the scanner, Dr. Hadaschik pointed out. That requires a rare level of cooperation between the specialties of urology and radiology — something that will be hard to replicate elsewhere.

"In-bore is something that urologists are a bit afraid of because then the patient is diagnosed and counseled by a radiologist, and might be sent to radiation oncology. It's a territorial thing. In-bore targeted biopsies are always done within the magnet, and the magnet belongs to the radiologist." Although the Australian team actually involved a urologist in the performance of the in-bore biopsies, "this is nothing that will be applicable to a huge number of men," Dr. Hadaschik explained. A more likely shift will be toward image fusion, where urologists use MRI and ultrasound information to guide their own procedures," he said.

TRUS Just as Good?

In the study presented by Dr. Hiester, which suggested that TRUS-guided biopsy can be as good if done well, 128 men (mean age, 66 years) with an elevated PSA level (median, 6.7 ng/mL) were prospectively analyzed using both mpMRI and TRUS. Median prostate volume was 55 mL.

All patients underwent a diagnostic 3 Tesla mpMRI; prostate lesions were defined using the PI-RADS. All lesions were histologically verified with targeted in-bore MRI-guided biopsy (2 biopsies per lesion and a mean of 5 biopsies) and then a standardized TRUS-guided 12-core biopsy.

The prostate cancer detection rate with the 2 biopsies combined was 60.9%. When used alone, the MRI and TRUS biopsies had an identical detection rate — 53.1% — and a similar ability to detect clinically significant intermediate- and high-risk cancer (22 vs 23 patients). "We did not find any significant difference in Gleason score between the 2 methods," Dr. Hiester said.

More than half of the lesions missed by TRUS-guided biopsy were located in the interior part of the prostate gland, he added.

"Multiparametric MRI is certainly a useful improvement in the diagnosis of prostate cancer," Dr. Hiester told Medscape Medical News. "But there are still randomized comparative studies missing to really prove the advantage of the MRI-guided in-bore biopsy. I am convinced that mpMRI is the way ahead in the diagnosis of prostate cancer, but it is too early to declare it as a standard procedure in biopsy-naive patients."

However, "one of the big advantages" of the MRI-guided approach is that the yield per biopsy is greater than with TRUS, he said. The percent of positive biopsies was significantly greater with MRI than with TRUS (35.4% vs 14.4%; < .0001), and the mean number of biopsy cores per patient was smaller with MRI (5.3 vs 12.0). For tumors with equivalent Gleason grading, the mean tumor infiltration of biopsy cores was also greater with MRI biopsy (61.4% vs 48.4%; < .0001).

Dr. Hadaschik told Medscape Medical News that this study had an unusually good TRUS detection rate.

"There were very dedicated urologists doing the 12-core and there were fairly small prostates, so there was relatively good geometrical distribution and they sampled most of the prostate," he said. "And if you do the targeted MRI biopsy first and then do the 12-core TRUS biopsy, even with the urologist blinded to the MRI findings, you can still see the biopsy track, which would influence where you sample. I think in this negative study, the bias is toward an improved control arm."

Dr. Hadaschik said there is an unmistakable shift in urology toward targeted approaches.

"There is a very strong community within urology that believes we are not doing a good job at the moment by overdiagnosing many men with indolent tumors. MRI is by far the best tool to help us move away from that. As a community, we must reduce the number of systematic biopsies. To do that, we must plan multi-institution randomized trials."

Dr. Pokorny, Dr. Hadaschik, and Dr. Hiester have disclosed no relevant financial relationships.

European Association of Urology (EAU) 29th Annual Congress: Abstracts 948 and 947. Presented April 14, 2014.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.