By: Dan Sperling, MD

As with so many medical technologies, multiparametric magnetic resonance imaging (mpMRI or functional MRI) had its earliest applications for prostate cancer (PCa) at a handful of scattered locations around the globe.  Gradually, the number of centers and users grew, and professional publications and dialogue enriched knowledge about what it could and could not do.

A new article by a group of Australian academics who surveyed published literature on mpMRI presents a thoughtful summary of the role of prostate mpMRI today, and projects its value into the future.[i]

The article opens with an ongoing PCa problem: poor matching of treatment to disease. Conventional PSA-based screening fosters the overdetection and overtreatment of insignificant PCa; many patients are sent for radical treatments that are economically burdensome to the medical care system, and risks leaving them with impaired urinary and sexual function. On the other hand, both active surveillance (AS) and focal therapies carry the risk of undertreating an undetected but highly aggressive cell line.

This situation is hardly remedied by ultrasound-guided biopsy techniques (both transrectal and transperineal) that are blind to tumor location and therefore quite literally hit-or-miss. Even when positive for PCa, the authors point out that tissue analysis underestimates the Gleason grade up to 30% of the time. This is the first area in which mpMRI has an important part to play. “In the best hands, mpMRI of the prostate has a specificity approaching 90% and a negative predictive value of around 85%,” according to the authors. This means that when scanned and interpreted by experts, what shows up as prostate cancer IS cancer 90% of the time, and when the MRI does not reveal cancer it is correct 85% of the time. Compare those accuracy numbers with TRUS biopsy that finds cancer about 70% of the time, and the value of mpMRI is evident—with the qualification that the magnet, software, and interpreter are all top of the line.

The next section of the article explains how each of four parameters detects abnormalities, and how the parameters may be combined to increase the accuracy of the scan results.

• T2-weighted imaging reveals the different zones and capsule anatomy of the prostate. By itself, it may detect PCa in the peripheral zone, but in many cases, especially in the transitional zone, the results may be ambiguous.
• Diffusion weighted imaging (DWI) distinguishes PCa from healthy cells based on the natural movement (diffusion) of water molecules within tissues. Because the density of tumor cells restricts water motion, a calculation of the apparent diffusion coefficient, or ADC, can be made. The lower the ADC value, the greater the restriction and therefore likelihood of cancer. DWI sequences can be varied according to the “b value” (strength of the magnetic field) so that at higher b values the sensitive to water diffusion is increased while anatomic detail is filtered out. While suspicious areas show up as dark regions on an ADC map, images with the highest b values show the same regions as bright spots. Therefore, it’s worth the scanning time to run ADC and high b value DWI sequences.
• Dynamic contrast enhanced (DCE) MRI relies on the intravenous injection of a contrast agent that can reveal the abnormal blood flow associated with cancerous tumors. According to the article, “It has been shown to be able to detect significant disease in up to 93% of cases.”
• Magnetic resonance spectroscopy (MRI-S) is sensitive to the presence of biochemical substances called metabolites that PCa cells particularly need to fuel their growth. While it was more widely used when first introduced, it adds more time to the scan and some authors have demonstrated it does not really add to the accuracy of the MRI results. It is much less commonly used today in the U.S.

Because each of these parameters tunes into different tissue or blood flow functions, mpMRI is often called functional MRI, a term that many radiologists prefer. In any case, combining the functional parameters of T2W, DWI and DCE allow the most confident interpretation of PCa since one of them may be weaker than the other two depending on the size, location and aggressiveness of the tumor. This combination of more than one parameter is what led to use of the term multiparametric MRI (meaning MRI with multiple parameters used in combination).

The authors point out that since mpMRI has been implemented in an ad hoc (as developed and used at the time) fashion, efforts are being made to standardize risk assessment based on images. A 5-point scale that originated with breast cancer imaging, called BIRADS (breast imaging-reporting and data system) has been adapted for prostate cancer imaging and thus termed PIRADS. The higher the number assigned to each parameter, the greater the likelihood of malignancy. When the values for T2W, DWI and DCE are totaled, the scores will range from 3-15; if MRI-S is included, the range will be from 4-20. While there is not yet universal consensus on a scoring system, PIRADS appears to be the most favored.

The current contribution of mpMRI is established by comparing a patient’s imaging results with that same person’s subsequent biopsy results (pathology) and/or prostatectomy results (histopathology) if that was the treatment choice. For patients whose biopsy was negative but blood tests continued to be suspicious, at least one study has shown that the combination of T2W and DWI had a “higher positive rate than any individual sequence when the biopsy result was negative.” Although the data varies, the concordance between mpMRI results and post-prostatectomy specimens has been consistently significant for localized PCa. However, some studies have shown the mpMRI was a poor predictor of extracapsular extension (tumor that has penetrated beyond the prostate capsule) and seminal vesicle invasion. On the other hand, combining all parameters (T2W, DWI and DCE) with basic T1 imaging has been found effective in detecting extracapsular extension by other researchers. Likewise, mpMRI images correlate well with biopsy, especially when the needles are selectively guided to the suspicious areas as shown on the scans.

Finally, the authors specify present and future contributions of prostate mpMRI, including:

• Detecting PCa in patients with one or more previous negative biopsies
• Detecting recurrence in post-prostatectomy and radiation patients
• Detecting significant PCa to assist in treatment planning
• Targeting biopsies to improve detection rates, enhance diagnostic accuracy and eliminate unnecessary biopsies
• Improving the safety of AS by monitoring with repeat imaging and implementing targeted biopsy to confirm suspected disease progression.

Thus, the authors conclude that prostate mpMRI “is a promising technology within PC management.” They call for continued studies in order to produce ever more robust data, and they caution that at this time it should not be seen as a replacement for diagnosis by taking tissue samples. They point out that the cost and limited availability of well-equipped centers prevent mpMRI being used for wide screening. And while the article does not cover the role of mpMRI in qualifying patients for focal treatment, many of the observations in it imply its merits in this regard. The bottom line is that mpMRI of the prostate, used in conjunction with other screening and diagnostic methods, has a powerful role “as a useful tool in PC diagnosis and management as well as a reliable means of assessing men in the context of active surveillance.”

[i] Katelaris N, Bolton D, Weerakoon M et al. Current role of multiparametric magnetic resonance imaging in the management of prostate cancer. Korean J Urol 2015;56:337-45.