Comparison of Multiparametric MRI Scoring Systems and the Impact on Cancer Detection in Patients Undergoing MR US Fusion Guided Prostate Biopsies

Abstract

IntroductionMultiple scoring systems have been proposed for prostate MRI reporting. We sought to review the clinical impact of the new Prostate Imaging Reporting and Data System v2 (PI-RADS) and compare those results to our proposed Simplified Qualitative System (SQS) score with respect to detection of prostate cancers and clinically significant prostate cancers.MethodsAll patients who underwent multiparametric prostate MRI (mpMRI) had their images interpreted using PI-RADS v1 and SQS score. PI-RADS v2 was calculated from prospectively collected data points. Patients with positive mpMRIs were then referred by their urologists for enrollment in an IRB-approved prospective phase III trial of mpMRI-Ultrasound (MR/TRUS) fusion biopsy of suspicious lesions. Standard 12-core biopsy was performed at the same setting. Clinical data were collected prospectively.Results1060 patients were imaged using mpMRI at our institution during the study period. 341 participants were then referred to the trial. 312 participants underwent MR/TRUS fusion biopsy of 452 lesions and were included in the analysis. 202 participants had biopsy-proven cancer (64.7%) and 206 (45.6%) lesions were positive for cancer. Distribution of cancer detected at each score produced a Gaussian distribution for SQS while PI-RADS demonstrates a negatively skewed curve with 82.1% of cases being scored as a 4 or 5. Patient-level data demonstrated AUC of 0.702 (95% CI 0.65 to 0.73) for PI-RADS and 0.762 (95% CI 0.72 to 0.81) for SQS (p< 0.0001) with respect to the detection of prostate cancer. The analysis for clinically significant prostate cancer at a per lesion level resulted in an AUC of 0.725 (95% CI 0.69 to 0.76) and 0.829 (95% CI 0.79 to 0.87) for the PI-RADS and SQS score, respectively (p< 0.0001).ConclusionsmpMRI is a useful tool in the workup of patients at risk for prostate cancer, and serves as a platform to guide further evaluation with MR/TRUS fusion biopsy. SQS score provided a more normal distribution of scores and yielded a higher AUC than PI-RADS v2. However until our findings are validated, we recommend reporting of detailed sequence-specific findings. This will allow for prospectively collected data to be utilized in determining the impact of ongoing changes to these scoring systems as our understanding of mpMRI interpretation evolves.

Differentiation of prostate cancer lesions in the Transition Zone by diffusion-weighted MRI

“…[25], [26], [9]. MR/TRUS fusion-biopsy is routinely used in clinical practice to integrate into biopsy planning information gained by MRI [27].…”

An initial negative round of targeted biopsies in men with highly suspicious multiparametric magnetic resonance findings does not exclude clinically significant prostate cancer—Preliminary experience

“…Refinements in mpMRI, such as the use of scoring systems that ascribe a likelihood of clinical significance to the MRI findings, either by using a Likert scale or PIRADS, support an ability to correlate highly suspicious appearing lesions with higher risk cancers [4-11]. This has led some to suggest that a negative MRI-TRUS fusion targeted biopsy or even a negative mpMRI might be a sufficient means to exclude significant disease[17].…”

“…The degree of suspicion on multiparametric MRI (mpMRI) is the strongest predictor of a positive targeted biopsy with positive rates of 70-90% when the MRI findings are highly suspicious for PCa (i.e., PI-RADS or Likert scores of 5) [4-11]. Thus, a negative targeted biopsy of a highly suspicious lesion on mpMRI must be interpreted with caution and additional actions (e.g., repeat biopsy, imaging follow-up) may be required depending on the clinical context.…”

Magnetic Resonance Imaging-Ultrasound Fusion-Guided Prostate Biopsy: Review of Technology, Techniques, and Outcomes

“…A recent publication reported decrease in diagnostic accuracy of PIRADSv2 compared to 5 point Likert scale based on the ESUR guidelines from 2012 with each patient acting as their own control (66). Standardized reporting is essential, however interpretation is subjective and proposed scoring systems should reflect appropriate risk stratification with respect to the presence of CSPCa (67).…”

“…Small differences in interpretation can alter outcomes of fusion biopsy devices and make comparing systems between institutions and historical data almost impossible. The authors advise centers to record granular sequence specific data to allow for future analyses (66). …”