Researchers Look to MRI and Biomarkers to Help Improve Detection of Aggressive Prostate Cancers

A patient’s wife once told urology surgeon Peter Pinto, MD, that there was no way she’d agree to have a dozen biopsy needles stuck blindly into her breast just because a blood test suggested she might have cancer.

But that’s exactly what happens with many men whose cancer screening test reveals an elevated level of prostate-specific antigen, or PSA, in their blood. Although mammography provides images of suspicious breast lesions to guide biopsy needles, prostate cancer is the only type of solid organ tumor that is usually diagnosed sight unseen with hit-or-miss tissue biopsies.

Imaging the prostate simply doesn’t offer enough information for disease detection. Transrectal ultrasound, routinely used during prostate biopsies, provides images of the gland but not of the tumors inside of it. For decades, urologists have used a systematic but blind approach to sample prostate tissue in men with an elevated PSA level and no palpable lesion, taking up to 18 core needle biopsies in scattered sections of the organ. But they can’t be sure that they haven’t missed an aggressive tumor hidden in the 99% of tissue that’s not biopsied, and they sometimes end up sticking needles into men who have no tumors, only an elevated PSA level.

Before the PSA test became available nearly 30 years ago, physicians didn’t look for prostate tumors in asymptomatic men, let alone treat them, and autopsy studies found that some died with indolent cancers that never caused any harm. But the simple blood test has led to what many consider the overdiagnosis and overtreatment of low-risk prostate cancers, which can leave men impotent, incontinent, or both.

The PSA test’s imperfections have spurred scientists to search for new imaging approaches and biomarkers to improve the diagnosis and management of prostate cancer, expected to cause 27 540 deaths in the United States this year (http://bit.ly/1IKzCOd). It’s not clear whether these tools could ever be used as primary screens for prostate cancer, but they hold promise as a next step after PSA screening suggests a man might have the disease. The goal: to better detect high-risk tumors and distinguish them from low-risk tumors, thereby optimizing the chance that men who are most likely to benefit from biopsies or treatments will be the most likely to get them.

One approach combines transrectal ultrasound’s ease of use with the wealth of information magnetic resonance imaging (MRI) can provide. Thanks to a recent advance called multiparametric MRI, specially trained radiologists can gauge the aggressiveness of a prostate lesion not only by how it looks but also by how tightly its cells are packed, how blood flows through it, and its chemical makeup.

For now, a urologist must biopsy the lesion in question to confirm the radiologist’s findings, but performing the procedure while a man is in the MRI tube is cumbersome, time-consuming, and expensive. Instead, urologists can electronically fuse MRIs with real-time transrectal ultrasound in their office and then aim biopsy needles directly at lesions.

The US Food and Drug Administration has approved a handful of systems that fuse prostate MRIs with transrectal ultrasound. For the most part, however, the technology remains the bailiwick of research centers trying to pinpoint its place in the prostate cancer diagnostic continuum.

In a new prospective study published recently in JAMA, Pinto, a faculty member in the urologic oncology branch at the National Cancer Institute (NCI), and his colleagues analyzed data from 1003 men referred for an elevated PSA level or abnormal digital rectal examination. The investigators found that targeted biopsy using MR/ultrasound fusion diagnosed more high-risk cancers and fewer low-risk cancers than conventional biopsy (Siddiqui MM et al. JAMA. 2015;313[4]:390-397). Pinto’s team used a platform called UroNav, the result of nearly a decade’s research conducted mainly at the NCI, although the NCI scientists have no financial stake in the product.

The study began enrolling men in 2007, but “this paper does not have the long-term follow-up to see improvements in prostate cancer–specific mortality or overall mortality. For prostate cancer, that often takes years and decades,” Pinto says. “We are following these patients and look forward to future publications.”

At the University of California, Los Angeles (UCLA), virtually all men with a PSA level between 4 and 10 get a multiparametric MRI before proceeding to biopsy, urology professor Leonard Marks, MD, says. “It does raise the price of poker when you do the MRI before biopsy,” Marks says. But without an MRI, you’d have to biopsy 4 men to find 1 cancer, and only half of those cancers would be potentially life-threatening, he says. “A lot of these men undergo biopsies over and over again.”

Still, those blind biopsies miss aggressive tumors, Marks says, citing the case of a UCLA dental school faculty member who, because of an elevated PSA level, underwent 8 systematic biopsy procedures, all negative, over 10 years. It wasn’t until he had a multiparametric MRI that an aggressive tumor was detected and, with MR/ultrasound fusion, biopsied, resulting in a prostatectomy.

Even if nothing suspicious is seen on the MRI, Marks says, he will still do a systematic biopsy procedure in men with an elevated PSA level using an MR/ultrasound fusion device called Artemis. Artemis allows him to map the biopsy for future reference. “Some important tumors, maybe 20%, are ‘MRI invisible,’ so it’s important at least once to do the mapping biopsy,” he says.

Urologist Geoffrey Sonn, MD, who trained with Marks at UCLA and is now on the faculty at Stanford, notes that for now, “we don’t know for sure what a negative or normal MRI means,” which is why he also performs a “mapping biopsy” if no lesion is seen. However, Sonn says, the tumors that multiparametric MRI is most likely to miss are small, low-grade cancers. “We think that the vast majority of the ones we want to find are visible on the MRI.”

Still, Marks says, “MRI is never going to be as easy as a PSA blood test. MRI is the next level. What we need is a better screening test.” Concern about overtreatment and overdiagnosis led the US Preventive Services Task Force in 2012 to advise against screening average-risk men with no prostate cancer symptoms (http://bit.ly/1JfANpa). And in 2013, the American Urological Association (http://bit.ly/1KK6WJ3) and the American College of Physicians (http://bit.ly/1utrosM) published prostate cancer screening guidelines recommending that physicians discuss the risks and benefits of screening with patients before proceeding.

As with multiparametric MRI, commercially available biomarker tests, such as those that measure various forms of PSA in the blood (http://bit.ly/1umbhrn), can be used after screening detects an elevated PSA level.

One biomarker entering the prostate cancer screening arena is prostate cancer antigen 3 (PCA3), a gene that is overexpressed in prostate cancers and can be measured in urine (http://bit.ly/1J8I3Vr). A recent NCI study at 11 medical centers concluded that the use of PCA3 testing among men screened with PSA could reduce the burden of prostate biopsies in men undergoing a repeat biopsy (Wei JT et al. Clin Oncol. 2014;32[36]:4066-4072). A high PCA3 score in men who haven’t had a biopsy significantly increases the likelihood that an initial biopsy will identify cancer, the researchers found.

Additional advances have suggested the potential diagnostic and prognostic utility of multipanel urinary biomarker screens that measure PCA3 along with the TMPRSS2:ERG fusion transcript, which has been reported to be present in more than 50% of PSA-screened prostate cancers (Tomlins SA et al. Sci Transl Med. 2011;3[94]:94ra72; Leyten GH et al. Eur Urol. 2014;65[3]:534-542).

However, many of these tests are not currently in widespread clinical use, and studies are ongoing to further validate their clinical utility.

Some urologists predict that multiparametric MRI could be useful for men who have opted for active surveillance—sometimes called “watchful waiting”—instead of treatment for what appears to be indolent prostate cancer. “In the ideal world, every single patient going on surveillance would have an MRI,” says Klotz, coauthor of a recent review article about using MRI in active prostate cancer surveillance (Schoots IG et al. Eur Urol. doi:10.1016/j.eururo.2014.10.050 [published online November 15, 2014]).

About 1 in 4 men who chose active surveillance actually harbor a more aggressive cancer than their original systematic biopsy suggested, he says. Over a period of 15 years, perhaps 1 in 50 men on active surveillance die because their aggressive cancer wasn’t detected soon enough, Klotz says.

Klotz is the principal investigator in a multisite clinical trial seeking to determine whether MR/ultrasound fusion biopsy can improve selection of patients eligible for active surveillance (http://1.usa.gov/1yrhbNd). He says the study has nearly finished enrolling 300 active surveillance patients who are being randomized to receive either a conventional systematic biopsy or an MR/ultrasound fusion targeted biopsy a year after they enter the study, so the results won’t be available until 2016.

“I do have patients for whom I am more likely to recommend active surveillance …just because I’m reassured by having an MRI that doesn’t look that abnormal,” Sonn says.

Johns Hopkins offers active surveillance patients the opportunity to have an MRI during their first year in the program, says H. Ballentine Carter, MD, director of the adult urology division and codirector of the active surveillance program. “We offer it as a way of decreasing the risk that we’re misclassifying people,” Carter says. “We try to emphasize to them that we’re still learning about MRI. We don’t have all the answers. We don’t know that this test will provide any outcomes that will be meaningful to them.”

Although he doesn’t think that MRI adds much for men who meet all of the criteria for Hopkins’ active surveillance program, more than 90% ask for one, Carter says. “It’s another piece of information …to help ensure that we’re monitoring the right person.” If nothing suspicious is seen on the MRI, doctors will still perform a systematic biopsy. “Our hypothesis is that men who have nothing suspicious on an MRI are extremely unlikely to have something serious later on,” he says.

Carter and the other physicians expressed concern that marketing practices for MR/ultrasound fusion technology are ahead of the science. For example, search for “UroNav” or “Artemis” on Google, and you’ll find several community urology practices touting that they have it.

“That kills me,” Pinto says. “All we can do is continue to publish and continue to train.”

Radiologists who are inexperienced in interpreting prostate MRIs tend to see tumors around every corner, Klotz says. “Reading these multiparametric MRIs is a fairly specialized skill, and that skill is just diffusing out there.”

Still, he predicts, multiparametric MRI might one day supplant the need for prostate biopsies, as computed tomographic scans have in the treatment of kidney cancer. Says Klotz: “The imaging is so powerful.”