Initial T1-weighted images are performed first to determine if hemorrhage is present in the prostate. As such, most experts recommend waiting 3-8 weeks after a prostate biopsy to decrease artifact associated with hemorrhage from the biopsy.2 Subsequently, T2-weighted images provide anatomic configuration of the prostate gland: the normal peripheral zone appears as areas of high signal intensity, whereas areas of low signal intensity may represent prostate cancer, prostatitis, BPH, etc. T2-weighted images also provide information regarding extraprostatic extension (EPE) or seminal vesical invasion (SVI), which are represented by areas of low signal intensity. DWI assess the diffusion of water within the magnetic field—the closer the cells are together (ie. for a prostate cancer nodule), the lower the motion of water, which leads to a high signal intensity in this phase. The DCE phase is a T1-weighed image with gadolinium-based contrast, which assesses vascular permeability of the prostate over a period of typically 5-10 minutes. Importantly, the combination of T2, DCE, and DWI phases yields both a NPV and PPV of >90%.3
The objective of this article is to focus on indications for mpMRI use in the localized prostate cancer setting, specifically exploring its use before prostate biopsy, after a negative biopsy, on active surveillance, and prior to radical prostatectomy for surgical planning purposes.
Before Prostate BiopsyUntil recently, the utilization of mpMRI as a “triage test” prior to transrectal ultrasound (TRUS)-guided prostate biopsy was a contentious topic, relying on single centers suggesting that such an approach may decrease unnecessary biopsies.5,6 However, in 2017, the PROMIS trial provided level 1b evidence for utilizing mpMRI prior to TRUS-guided biopsy among men with elevated PSA.7 PROMIS was a multi-center, paired-cohort study to assess the diagnostic accuracy of mpMRI and TRUS biopsy against a gold-standard reference template mapping biopsy. Men were included (n=576) if they had a PSA <15 ng/ml and no history of the previous biopsy. On mapping biopsy, 71% of men had cancer, including 40% with clinically significant prostate cancer (Gleason score ≥4+3 or maximum cancer length ≥6 mm). For clinically significant disease, mpMRI was more sensitive (93%) than TRUS-biopsy (48%), albeit less specific (41% for mpMRI; 96% for TRUS-biopsy). Based on these data, a triage mpMRI would allow 25% of men to safely avoid a prostate biopsy, while at the same time reducing detection of clinically insignificant prostate cancer. Importantly, secondary to the poor specificity and positive predictive value, this study does not suggest that mpMRI should replace prostate biopsy and that men with suspicious lesions should still have histologic confirmation of prostate cancer.
Shortly after PROMIS was published, PRECISION reported results of their trial which assigned 500 men with a clinical suspicion of prostate cancer who had not previously undergone a prostate biopsy to undergo MRI with or without a targeted biopsy vs standard TRUS-guided biopsy.8 Men in the MRI group underwent a targeted biopsy if there was a suspicion of prostate cancer on imaging and did not undergo a biopsy if the MRI was negative. The primary outcome for this randomized clinical trial was a diagnosis of clinically significant prostate cancer. In the MRI-targeted biopsy group, 28% had a negative MRI and thus no biopsy. Among men undergoing targeted biopsy, 38% had clinically significant cancer, compared to 26% in the TRUS-guided biopsy group (p=0.005). Furthermore, fewer men in the MRI-targeted biopsy group had clinically insignificant prostate cancer compared to the TRUS-guided biopsy group.
Since the publication of these two trials, debate regarding the implementation of mpMRI prior to biopsy has ensued. Detractors have mentioned that the negative predictive value of mpMRI in PROMIS for detecting Gleason grade group ≥2 was only 76%, albeit no grade group ≥3 were missed on mpMRI.7 mpMRI prior to prostate biopsy has already been widely accepted in the UK and Australia where mpMRI is reimbursed in this setting.9 There is evidence to suggest that men with a negative mpMRI should not be biopsied unless caveats such as a strong family history, abnormal digital rectal exam, or BRCA mutation are present. At present, experts have pressed for a new paradigm for prostate cancer detection in which an abnormal mpMRI should have a targeted biopsy performed; if the mpMRI is negative then a routine follow-up protocol should be employed.9 It is noteworthy to mention that widespread dissemination will undoubtedly rely on (i) affordability/reimbursement of mpMRI, (ii) quality of the mpMRI, and (iii) skill of the radiologist.
After Negative Prostate BiopsyIt is generally accepted that men with a history of negative TRUS-guided biopsy and persistently elevated or increasing PSA should undergo a mpMRI prior to consideration of a second (or in some cases third or fourth) prostate biopsy. In a study of 265 patients with a PSA >4.0 ng/ml and one negative TRUS-biopsy mpMRI detected prostate cancer in 41% of men, including 87% with clinically significant prostate cancer.10
mpMRI has been shown in the previous negative biopsy setting to detect tumors in up to 40% of cases, often in the anterior region of the prostate.11-14 An advantage of mpMRI fusion biopsy in patients with prior negative biopsy is the ability of MRI to identify suspicious lesions in areas not normally sampled by standard TRUS-guided biopsy, specifically the anterior and apical parts of the prostate. Thus, the benefit of fusion biopsy is particularly accentuated in the prior negative biopsy cohort. Furthermore, a study by Kongnyuy et al.15 suggested that there may be racial differences with regards to anterior tumors. In a cohort of 195 African-American men matched 1:1 to white men undergoing mpMRI, 47.7% of African-American men had anterior prostate lesions. Amongst these men, a history of prior negative biopsy was significantly associated with an anterior prostate lesion (OR 1.81, 95%CI 1.03-3.20). Despite an overall higher cancer detection rate among African American than white men, the presence of anterior prostate lesions and lesions harboring clinically significant cancer were not different between races.
On Active SurveillanceOver the last decade, adoption of active surveillance as a management strategy for men with clinically low-risk prostate cancer has appropriately continued to increase. However, until recently, the utilization of mpMRI in active surveillance management has been somewhat discretionary and clinician dependent.16
Earlier this year the European Association of Urology (EAU) released a position statement for active surveillance, which included 10 recommendation statements;17 the 3rd statement assessed “use and timing of MRI in active surveillance.” mpMRI can be used to increase clinically significant cancer detection, thus ensuring men are appropriately included in surveillance regimens and those with potentially threatening disease can have appropriate and timely intervention. The statement recommends that mpMRI can be performed at several time points during active surveillance:
- At the time of initial diagnosis – the EAU statement recommends that men diagnosed with a low-risk disease without a prior mpMRI should under a mpMRI prior to enrolment to ensure no significant disease was missed on initial biopsy. In cases of initial targeted mpMRI biopsy, both targeted and systematic biopsies should be performed.17 In addition to the excitement generated by PROMIS 7 and PRECISION,8 a recent systematic review assessed the role of mpMRI among active surveillance patients, noting that a lesion suspicious for prostate cancer was found in nearly two-thirds of men otherwise suitable for surveillance.18
- Before confirmatory biopsy – the EAU statement recommends that a mpMRI be performed before the confirmatory biopsy, within 12 months from initial diagnosis, and to include targeted and systematic biopsies.17 The rate of reclassification after targeted biopsies among men on active surveillance without a prior mpMRI may be as high as 22%.18-20 A recent publication assessed the value of serial mpMRI imaging among 111 men on active surveillance with > 1-year of follow-up, noting that among 33 reclassifications after one year, 55% were reclassified on only TRUS-guided biopsy.21 As such, the value of serial mpMRI in active surveillance algorithms remains unclear.
- During follow-up – the EAU statement does not support the use of solely using mpMRI instead of repeat biopsy in active surveillance follow-up.17 As mentioned, the use of serial mpMRIs over long-term follow-up is not currently recommended, however it may be used in situations where a targeted lesion is being followed. This is an area of great research interest considering that institutional studies with vast experience with mpMRI suggest that mpMRI supplanting follow-up biopsies is safe and feasible.22
Before Radical ProstatectomyWith improved mpMRI technology has come an interest in more precise clinical staging of localized prostate cancer, particularly before performing radical prostatectomy. Ultimately, the patient and urologist are concerned about the risk of EPE preoperatively, which dictates the degree of nerve-sparing performed at the time of radical prostatectomy. Somford et al.24 assessed mpMRI images among 183 men to determine the positive and negative predictive values of mpMRI for EPE at radical prostatectomy for different prostate cancer risk groups. The overall prevalence of EPE at radical prostatectomy was 49.7% (24.7% low-risk; 77.1% high-risk) – the overall staging sensitivity was 58.2%, specificity was 89.1%, positive predictive value was 84.1% and a negative predictive value was 68.3%. The positive predictive value was best in the high-risk cohort (88.8%) and a negative predictive value was best in the low-risk cohort (87.7%).
Data regarding whether preoperative imaging influences surgical planning is limited. However, Schiavina et al.25 assessed the impact of mpMRI on preoperative decision making among 137 patients planned for radical prostatectomy who underwent mpMRI. They found that mpMRI changes robotic surgeon’s initial surgical plan with regards to the degree of nerve-sparing in nearly half of patients. Interestingly, there was an equal alteration in surgical planning when considering a more aggressive (to less aggressive) and less aggressive (to more aggressive) preliminary plan. Although the above results for EPE prediction and tailored surgical planning are encouraging, this degree of advanced mpMRI interpretation should be reserved for expert radiologists where sensitivities and specificities for predicting EPE are typically both >80%.26