What Urologists Need to Know: mpMRI for the Detection of Clinically Significant Prostate Cancer Journal Club - Christopher Wallis & Zachary Klaassen

January 12, 2023

In this UroToday Journal Club Christopher Wallis and Zachary Klaassen highlight the European urology publication titled "Multiparametric Magnetic Resonance Imaging for the Detection of Clinically Significant Prostate Cancer: What Urologists Need to Know. Part 2: Interpretation." It is generally felt that MRI is superior to other imaging techniques, including CT scans, for soft tissue assessment. This data focuses on how we may reduce the interpretation variability and achieve optimal accuracy in assessing prostate MRI for optimizing its clinical use.


Christopher J.D. Wallis, MD, Ph.D., Assistant Professor in the Division of Urology at the University of Toronto.

Zachary Klaassen, MD, MSc, Urologic Oncologist, Assistant Professor Surgery/Urology at the Medical College of Georgia at Augusta University, Georgia Cancer Center

Read the Full Video Transcript

Chris Wallis: Hello, and thank you for joining us. Today on UroToday, we're discussing the role of multiparametric MRI for prostate evaluation, and what urologists need to know. This is part two in a series, focused on the interpretation of MRI.

I'm Chris Wallis, assistant professor in the Division of Urology at the University of Toronto. With me today is Zach Klaassen, an assistant professor in the Division of Urology at the Medical College of Georgia.

You can see here, the citation for the publication, which underpins today's presentation.

By way of background, magnetic resonance imaging uses the interaction between radiofrequency pulses, a strong magnetic field, and body tissues, to obtain images of the body in three orthogonal planes. It is generally felt that MRI is superior to other imaging techniques, including CT scans, for soft tissue assessment.

In terms of multiparametric MRI evaluation of the prostate, the optimal technique includes three different sequences, including T2-weighted images, which are useful for depicting prostate anatomy, diffusion-weighted images, which display cellular density, and dynamic contrast-enhanced images, which show vascularity.

Clinical indications for multiparametric MRI of the prostate include, detection and localization of primary prostate cancer for guiding MRI directed biopsy, local prostate cancer staging, assessment of prostate cancer recurrence, and the role of active surveillance, and in planning local therapy.

The objective of today's presentation is to focus on how we may reduce the interpretation variability and achieve optimal accuracy in assessing prostate MRI for optimizing its clinical use.

T2-weighted imaging forms one of the three important components of a multiparametric prostate MRI. And T2-weighted images show the anatomic prostate zonal anatomy and the relation of the prostate to surrounding structures.

It's ideal for differentiating between the high-signal peripheral zone, mixed-signal transitional zone, and low-signal central zone. The transitional zone may be mixed as a result of BPH nodules, and we can see these highlighted in these figures here.

And so, these are all PI-RADS 1 images or BPH with a low probability of prostate cancer. And you can see well circumscribed nodules highlighted by these yellow arrows.

High-signal in a peripheral zone may be disrupted when there's a prostate cancer producing a lower signal area. However, prostate cancer may also have ISO signal areas, or non-focal mildly hypointense abnormalities, such that signal changes in T2-weighted images are not sensitive or highly specific for prostate cancer. Additionally, low-grade prostate cancer, or non-malignant conditions, can also have a low-signal intensity. And so, T2-weighted images cannot differentiate clinically significant prostate cancer from other conditions. However, it is notable that prostatitis typically has a wedge-shaped appearance that is somewhat characteristic. And you can see here, in these attached images, this PI-RADS 2 area, consistent with prostatitis, in the yellow circles, with these fairly well demarcated wedge like appearance.

So the diagnosis of clinically significant prostate cancer in the transition zone has greater challenges than the peripheral zone. And in particular, there's often ill-defined margins, and some focal homogenous T2 intermediate to low-signal, characterized as the quote erase charcoal drawing sign. There may be a variety of different shapes which are typically not discreet or well circumscribed. And there's this term, the disruption of organized chaos, describing the invasion of surrounding structures. And we can see this highlighted here in Panel D, with this disruption up here. And this is a PI-RADS 5 transition zone lesion.

In terms of diffusion weighted imaging, DWI allows further evaluation of tissue characteristics, and is important in differentiating clinically significant prostate cancer from other prostate conditions. Most important functional imaging is provided by these DWI sequences, because they correspond most strongly to histopathologic findings.

The normal peripheral zone has a high-signal which appears white due to high content of fluid-filled glandular structures, and the corresponding high velocity of water molecules. In contrast, when clinically significant prostate cancer replaces the healthy tissue, the high cellular density of the prostate cancer creates a low-signal on the ADC map, by restricting the diffusion of water molecules. Thus, lower ADC scores correlate with higher Gleason scores.

So on DCE images with contrast enhancement, there are a variety of conditions that may show contrast enhancement due to tumor angiogenesis, as well as vessel permeability, including both low-grade and high-grade prostate cancers, as well as other benign conditions. However, contrast-enhanced MRI is essential for the detection of local recurrence, and we need to use unenhanced T1-weighted imaging, to identify post biopsy hemorrhage, notable for a high T1 signal.

And when we put these together, the most widely adopted approaches, the PI-RADS system, this was originally developed in its initial form and subsequently updated to version 2 and version 2.1. It was developed by a joint steering committee from the European Society of Urogenital Radiology, the American College of Radiology, and the AdMeTech Foundation. The most recent version is version 2.1 and was released in 2019. The goal of PI-RADS version 2.1 is to simplify the assessment and reporting of prostate MRI with a goal of reducing interpretation variability. Is a risk assessment tool using a standardized evaluation method to predict the likelihood that clinically significant disease is present. As you can see highlight in this table here, PI-RADS version 2, categories one and two, are associated with a low or very low risk of clinically significant prostate cancer, raging between three and 12%. As the PI-RADS category increases up to category 5, the risk of clinically significant disease becomes very high, and depending on the report, is somewhere between 67 and 83%.

I'm now going to hand it over to Zach, to walk us through the details of PI-RADS version 2.1 in somewhat more detail.

Zach Klaassen: Thanks so much, Chris. So as Chris alluded to, version 2.1 of the PI-RADS system was released in 2019, and this emphasized the dominant role of diffusion-weighted imaging as the parameter for any suspicious lesion found in the peripheral zone, and T2-weighted imaging, in combination with diffusion-weighted imaging, in the transition zone.

As you can see here, in this figure provided by the authors, this really sort of delineates some of the nuances between detection in the peripheral zone and detection in the transition zone. You can see here in this red box, so this is where some of the details are further elaborated on. So a peripheral zone PI-RADS 3 lesion, if the DCE is negative, it stays as a PI-RADS 3. However, if the DCE is positive, this becomes a PI-RADS 4 lesion.

With regards to the transition zone, which highlights T2-weighted imaging, we see that really, the details are with regards to PI-RADS 2 and 3. So for PI-RADS 2, if DWI is less than three, this stays as a PI-RADS 2. If it's greater than four, or equal to four, this becomes a PI-RADS 3. And for transitional zone T2-weighted imaging, PI-RADS 3, if a DWI is less than four, it stays as a PI-RADS 3. However, if it's equal to five, this becomes a PI-RADS 4 lesion. So really, this figure is nice, because it highlights some of the other alternative sequences in the MRI that may either downstage or upstage the PI-RADS lesion, based off of these companion sequences in the MRI.

So how do we score these lesions? There's several important points here. So image quality must be assessed. If the quality of the MRIs in sufficient, either this is reported as a PI-RADS X, or the patient should be scheduled for additional imaging to obtain better imaging.

It's important that maximal prostate dimensions on T2-weighted imaging are measured in three perpendicular planes, and PSA identities also calculated at the same time.

So thirdly, before any lesions are scored or characterized, they need to be detected. PI-RADS is agnostic about the lesion detection method. And the recommendation for reading these or assessing these, is to look at the T2-weighted imaging, the high b-valued diffusion-weighted imaging, as well as early post-contrast enhancement images that are evaluated for any initial lesions that could represent clinically significant prostate cancer.

So the next several slides will determine some of the nuances to determining a PI-RADS 3 or 5 lesion.

So first, a focal mass in the peripheral zone with a low signal on ADC map, and a high signal on the high b-value DWI sequence, is either a PI-RADS 4 or 5, and this is really determined by the distinction of the cutoff of 15 millimeters. So greater than 15 millimeters is a five, less than is a four, or extra capsular extension.

The second example is if a TZ, or a transition zone lesion, has an erased charcoal appearance, and/or there is a disruption of organized chaos on T2-weighted imaging. Or if an anterior transition zone lesion has a lenticular shape, this is a PI-RADS 4 or 5 lesion.

Third, in the transition zone particularly, or partially encapsulated or circumscribed, or completely encapsulated nodules, as a T2-weighted image score of 2, with clearly restricted diffusion in a DWI score of four or five, this is a PI-RADS 3 lesion.

Fourth, a transition zone lesion with a T2-weighted image score of three, and a diffusion weighted image score of five, i.e., this is greater than 1.5 centimeters dimension, this is a categorized as a PI-RADS 4.

And finally, the dynamic contrast enhancement MRI sequence must be evaluated to see whether early enhancement in the peripheral zone matches with a wedge-shaped or diffuse intermediate signal ADC lesion, or a detected or undetected focal lesion. And if a lesion is in the peripheral zone, scored as three on DWI, which shows early focal enhancement, this is a PI-RADS 4 lesion.

So the next several slides, we'll look at some more MRI images, and go through some examples of additional PI-RADS imaging. So on this particular example, if we look at a panel A, this is a T2-weighted image showing extracapsular extension, with the circle. This is also confirmed of a sagittal coronal images, which are additional T2-weighted images. Additionally, as we see in panel B, this is a DWI low ADC value, a b 1400 high intensity lesion. And in Panel D, a dynamic contrast enhancement, early enhancement. This all correlates with a PI-RADS 5 lesion.

The second example, again, the T2-weighted axial image, from the apex in the midprostate, as well as the transition zone. We see this lesion listed here, confirmed on the sagittal in the coronal images in the T2-weighted sequence. Again on panel B, diffusion weighted imaging, low ADC value b 1400, a high intensity signal. However, on this one, we see dynamic contrast enhancement, no enhancement however, based on all of these criteria, this is still a PI-RADS 5 lesion.

In example three, we see a T2-weighted axial image with a midprostate small peripheral zone lesion, as you can see highlighted here, again on sagittal, in the T2-weighted imaging. In the coronal section, we see the image list in the circle. On diffusion-weighted imaging, no focal low-signal, but again, on b 1400, we see a high intensity lesion, and early enhancement on dynamic contrast enhancement imaging. And this is correlated with the PI-RADS score, based on the size criteria.

There is some limitation and challenges that the authors do mention, and these are important to discuss. And first this is, the studies show that multiparametric MRI as a triage test, is a cost effective approach. However, this is highly dependent on the quality of the multiparametric MRI and healthcare systems.

Secondly, there remains considerable variation in inter-reader reproducibility, but this is highly dependent on radiologists' experience and training.

And finally, an appropriate education program with quality control is needed for radiologists and urologists. And this is part of the reason that we're doing this series of multiparametric MRIs from the urologist perspective.

Future challenges include prebiopsy multivariable risk stratification using risk calculators, which includes PI-RADS, clinical data, pathology, and genomics, which needs to be developed and validated in future studies.

Secondly, guideline recommendations for clinical decision-making for each PI-RADS version 2.1 category, and subsequent biopsy results are needed.

And finally, to shorten examination time, biparametric MRI, which omits that dynamic contrast enhancement MRI phase, to exclude clinically significant prostate cancer in biopsy-naive men is increasingly being investigated, with promising initial results.

In conclusion, to enhance standardization, lesions must be scored using the PI-RADS assessment system.

Secondly, transition zone lesions that show erased charcoal or disruption of organized chaos, and peripheral zone lesions that are black on the ADC map, and white on the high b-valued DWI sequence, should be evaluated for a likelihood of clinically significant prostate cancer.

And finally, PI-RADS version 2.1 adds valuable information to other clinical data, and can be used to reliably exclude clinically significant prostate cancer, as well as indicate where MRI directed biopsies should be utilized.

We thank you very much for your attention, and we hope you enjoyed this UroToday discussion of a multiparametric MRI for urologists, and how to interpret these images. We hope you join us for subsequent part three and part four of this MRI series.