Multiparametric MRI (mpMRI) of the prostate combines anatomic magnetic resonance imaging (T1- and T2-weighted sequences) with at least one other functional sequence that characterizes healthy vs. malignant tissue. The functional sequences are called Diffusion Weighted Imaging (DWI), Dynamic Contrast Enhanced MRI (DCE-MRI) and MR Spectroscopy (MR-S). Pedler et al. (2015) have identified the current and potential roles of mpMRI in the detection, diagnosis, and treatment guidance of prostate cancer.[i] They note that a sufficiently powered magnet (preferably 3 Tesla) and experienced radiologist to interpret the scans are essential.
Numerous clinical studies have demonstrated that mpMRI is especially sensitive to significant prostate cancer (PCa). However, the definition of significance is still imprecise, though experts concur that it is sufficiently aggressive to require definitive treatment. One way to think of it is anything other than insignificant PCa, which is generally described as “(1) Gleason score < 6 without Gleason pattern 4 or 5, (2) organ-confined disease, and (3) tumour volume < 0.5 cm3 .”[ii]
Until recently, magnets in most hospitals and imaging centers were 1.5 Tesla machines. There is growing availability of 3 Tesla magnets as well as experienced radiological readers. It is not surprising, therefore, that the question arises: could mpMRI before prostate biopsy rule out the need for needle sampling of the gland? This issue is addressed in a recent study by Lamb et al. published in the journal Clinical Genitourinary Cancer.[iii] While acknowledging that the role of mpMRI in detecting clinically significant PCa is still controversial, they investigated the accuracy of mpMRI by comparing prebiopsy imaging with standard 12-core TRUS biopsy results. They included 173 patients during a 12-month period at their center who underwent mpMRI under a 1.5 Tesla magnet; the two sequences used in combination were T2-weighted and DWI, often considered the basic workhorse in detecting prostate tumors. The TRUS biopsies sampled from the apex, middle, and base of both sides of the gland.
Of the 173 patients, 128 (74.4%) were found on mpMRI to have a suspicious lesion, and 114 patients (66.3%) had a positive biopsy. When images were compared with biopsy-proven significant PCa, the results were:
mpMRI sensitivity 83.5%
mpMRI specificity 35.2%
mpMRI positive predictive value 55%
mpMRI negative predictive value 68.9%
The authors write: “A positive MRI scan was significantly associated with significant prostate cancer diagnosis… MRI detected 62 of 63 NCCN high-risk and 18 of 18 Gleason score 8 to 10 cases.” This indicates that MRI performed very well in detecting high-risk disease, but given that intermediate risk disease is also significant, the authors concluded that mpMRI “appears insufficient to avoid TRUS biopsy in all men clinically suspected to have prostate cancer.” They cautiously add that for those men at risk of biopsy complications, “… a negative MRI scan might be used to exclude high-risk disease.” The fact that the mpMRI studies were conducted on a less powerful magnet is disappointing, since it could be expected that a 3 Tesla magnet would have generated results with higher sensitivity and specificity for regions with intermediate-risk PCa.
Another recent study from a team of British experts poses a related question: What burden of prostate cancer can radiologists rule out on mpMRI?[iv] This is an interesting study because the imaging findings were compared with tissue samples taken by transperineal template prostate mapping biopsies (T-PMB), which provide more precise pathology information from each quadrant (sector) of the gland. (While not as thorough as a comparison with post-prostatectomy whole gland specimens, T-PMB is a superior mpMRI comparator over TRUS biopsies.) The purpose of the study was to identify the minimum tumor extent and aggressiveness at which mpMRI optimally performs.
The study enrolled 64 patients who first had mpMRI (index test) followed by T-PMB (reference test). For the image interpretation, each quadrant of the gland was evaluated by three radiologists on a 5-point scale:
1 – highly likely benign
2 – likely benign
3 – equivocal (could be either benign or malignant)
4 – likely malignant
5 – highly likely malignant.
According to the study, “The target condition indicating cancer on biopsies was varied by changing the maximum cancer core length (MCCL) and total cancer core length (TCCL) within each sector from 1 mm to 10 mm. The sensitivity, specificity, and positive (PPVs) and negative predictive values (NPVs) were calculated for each MCCL and TCCL. Gleason < 3+3 and Gleason > 3+4 cancers were analyzed separately.” By varying the target condition, it was possible to assess a cutoff point below which mpMRI did not perform well. Given that mpMRI generally has a high negative predictive value (the imaging correctly reflects the absence of significant PCa), the real question is whether it can reliably rule out need for a biopsy. In this study, 50% of the quadrants had biopsy-demonstrated PCa, of which 65% would be considered insignificant (Gleason < 3+3). Here are the mpMRI negative predictive value results:
MCCL of > 5 mm & TCCL of > 7 mm MCCL of > 5 mm & TCCL > 6 mm
Gleason < 3+3 NPV > 95%
Gleason > 3+4 NPV > 95%
In other words, the research team found that at a maximum needle core length of at least 5 mm (total core length at least 7 mm) mpMRI correlates well with insignificant PCa; and at maximum needle core length of at least 5 mm (total core length at least 6 mm) mpMRI correlates well with significant PCa. Based on their findings, the team concluded, “Multiparametric MRI may allow areas of the prostate which test negative to avoid biopsy.” They point out the need for further clinical trials to acquire more evidence.
These two studies point to the kind of critical questions that are being asked. Currently, it is not possible to establish a firm diagnosis of PCa without examining tissue samples, but given the number of men with biochemical (PSA) suspicion who are sent for TRUS biopsy only to learn they do not have PCa, mpMRI holds the promise of eliminating unnecessary biopsies. The second study in particular bears out that hope.
[i] Pedler K, Kitzing YX, Varol C, Arianayagam M. The current status of MRI in prostate cancer. Aust Fam Physician. 2015 Apr;44(4):225-30. Full article at http://www.racgp.org.au/afp/2015/april/the-current-status-of-mri-in-%E2%80%A8prostate-cancer/
[ii] Ploussard G, Epstein J, Montironi R et al. The contemporary concept of significant versus insignificant prostate cancer. Eur Urol. 2100;60:291-303.
[iii] Lamb BW, TAN WS, Rehman A, Nessa A et al. Is Prebiopsy MRI Good Enough to Avoid Prostate Biopsy? A Cohort Study Over a 1-Year Period. Clin Genitourin Cancer. 2015 Jun 30. pii: S1558-7673(15)00148-2. doi: 10.1016/j.clgc.2015.06.007. [Epub ahead of print]
[iv] Anastasiadis E, Charman SC, Arumainayagam N, Sohaib AS et al. What burden of prostate cancer can radiologists rule out on multiparametric magnetic resonance imaging? A sensitivity analysis based on varying the target condition in template prostate mapping biopsies. Urology. 2015 Aug 4. pii: S0090-4295(15)00479-3. doi: 10.1016/j.urology.2015.05.010. [Epub ahead of print]