By: Dan Sperling, MD

UroToday is an online new source for all things urological. Its mission to both physicians and patients is to provide “free, accurate, timely, evidence-based urology news and information…to access evidence-based diagnosis and treatment solutions from anywhere in the world.” In addition to posting daily urologic research news, a regular feature called “Beyond the Abstract” gives clinical contributors a forum for posting in-depth commentary on specific topics.

In March, 2015 UroToday carried a detailed essay by Sandeep Sankineni and Peter Choyke, both with the National Institutes of Health.[i] It was an opportunity for them to expand upon the brief abstract of their October 2014 review of the literature on functional MRI of the prostate.[ii]  They make an essential point: “If prostate MRI is to become cost-effective, its various components will likely need to be modified or eliminated to reflect patient-specific applications.” They make a case for differential use of prostate MRI rather applying a one-size-fits-all protocol.

Sankineni & Choyke summarize the evolution of the functional sequences (parameters) used most commonly today:

• T2 weighted imaging (T2 MRI) gives excellent anatomical detail. Originally, T2 MRI was done with an endorectal coil, primarily to stage tumors in terms of capsule and seminal vesicles
• Diffusion weighted imaging (DWI MRI) was incorporated in the mid-2000s. Coupled with “high b” values in addition to advances in magnetic gradients, DWI and T2 in combination has become the workhorse in prostate mpMRI.
• Dynamic contrast enhancement (DCE-MRI) came into use in the late 1990s, and today is used to round out information from T2 weighting and DWI.

MRI spectroscopy (MRI-S) was introduced in the mid-1990s, but is much less used today since it is not shown to add significantly to the accuracy of the above three parameters.[iii]

Historically, mpMRI scans typically involved all of the above functional sequences as they were developed, and until recently almost always involved an endorectal coil. According to the authors, “This created an unwieldy, uncomfortable and expensive hour-long scan. While multiparametric MRI proved its value in detecting and characterizing prostate cancer, it did so at a high cost.” To save scan time and cost, the authors therefore developed a simple chart to tailor which functional sequences are best applied to which cases (note absence of MRI-S which has no added value):

The authors note that since mpMRI is most sensitive to significant prostate tumors (Gleason > 3+4) there is always risk of missing small, insignificant lesions. However, continued screening with PSA and MRI will detect tumors on subsequent exams. Most importantly, any suspicious area detected by imaging is amenable to an MRI-guided targeted biopsy to confirm it is truly cancer.

The authors predict that adapting imaging to a customized “stripped-down” protocol on a case-by-case basis will result in a “more streamlined, less costly and less invasive” prostate MRI that will facilitate wider acceptance and adoption by the urologic community. This will help avoid unnecessary biopsies, increase diagnostic accuracy, and inform treatment choices including image-guided focal therapies.

[i] Reported March 10, 2015 by UroToday at http://www.urotoday.com/index.php?option=com_content&view=article&id=78992:functional-mri-in-the-detection-of-prostate-cancer-beyond-the-abstract-by-sandeep-sankineni-md-and-peter-choyke-md&catid=1134:prostate-cancer&Itemid=782&utm_source=newsletter_2505&utm_medium=email&utm_campaign=uroalerts-prostate-cancer-daily

[ii] Sankineni S, Osman M, Choyke PL. Functional MRI in prostate cancer detection. Biomed Res Int. 2014;2014:590638. doi: 10.1155/2014/590638. Epub 2014 Jul 23. Full article available at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4158139/

[iii] Platzek I, Borkowetz A, Toma M et al. Multiparametric Prostate Magnetic Resonance Imaging at 3 T: Failure of Magnetic Resonance Spectroscopy to Provide Added Value. J Comput Assist Tomogr. 2015 May 1. [Epub ahead of print]