Multiparametric MRI has rapidly emerged as a pre-biopsy triage test for ruling out significant prostate cancer (sPCa), avoiding the need for prostate biopsies. However, sPCas can be missed by MRI and additional predictors are, therefore, needed to separate men who require diagnostic biopsies from those who might safely avoid them. Furthermore, state-of-the-art multiparametric MRI is time-consuming and expensive and would place a significant financial and resource burden on any healthcare system if used before biopsy on all men. A simpler, more rapid biparametric MRI (bpMRI) method that uses fewer scan sequences, no intravenous contrast media, and maintains high diagnostic accuracy would decrease costs and could facilitate a more widespread clinical implementation of pre-biopsy prostate MRI.
In this paper, we assessed whether combining PSA density (PSAd) measurements (PSA divided by prostate volume) with bpMRI suspicion scores could improve the diagnostic accuracy and predictive values for detecting and ruling out significant Gleason score 7-10 PCa. In addition, we determined the best biopsy strategy and the proportion of men who could safely avoid biopsies based on bpMRI scores and PSAd thresholds. For this analysis, we used prospectively collected patient data from our institutional database and included 808 biopsy-naïve men with clinical suspicion of localized PCa (prostate-specific antigen <20 ng/mL, rectal examination <cT3). All patients underwent pre-biopsy bpMRI (T2- and diffusion-weighted imaging) followed by standard biopsies (all men) plus targeted biopsies of any suspicious (suspicion score 3–5) bpMRI findings. The detection rates of sPCa (Gleason score 7-10 PCa), the predictive values and the proportion of avoided biopsies were assessed for variable bpMRI scores and PSAd biopsy thresholds. The best biopsy strategy was determined using net benefit and decision curve analyses using combined biopsies from all men as the standard reference test.
The median age was 65 years and the median PSA level was 6.9 ng/mL. Overall, sPCa was detected in 283 of 808 men, and we found that PSAd significantly influenced the diagnostic yields of bpMRI in detecting and ruling out sPCa (Figure 1). Interestingly, the negative predictive value increased significantly from 83% to 95% (p = 0.002) for bpMRI scores of 1–2 (i.e., low suspicion MRI results) and from 53% to 93% (p < 0.001) for bpMRI scores of 3 (i.e., equivocal suspicion MRI results) when the PSAd was <0.15 ng/mL/cc. The best biopsy strategy, based on decision curve analysis of benefits and risks for biopsy thresholds ranging from 7.5% to 15%, was restricting biopsies to men with suspicious bpMRI results (i.e., scores of ≥4) or a PSAd ≥0.15 ng/mL/cc. This strategy decreased the number of men requiring biopsies by 41% (329/808) and decreased overdiagnoses of insPCas by 45% (79/177), while missing only 6% (17/289) of men with sPCas (the majority being Gleason score 3+4 PCas). The NPV of this strategy was high (95%), but the PPV was moderate (56%). Thus, although this strategy was apparently effective in ruling out sPCas (high negative predictive value), it was less effective at confirming the presence of sPCas, because 44% of men who tested positive would still undergo unnecessary biopsies.
Figure 1. Prostate cancer detection rates for all patients (N = 808) based on combined biopsy results stratified by bpMRI suspicion scores, PSA density, and PCa significance.
Overall, we concluded that a simple and rapid bpMRI together with PSAd may be used as a combined triage test to safely avoid prostate biopsies in biopsy-naïve men with clinical suspicion of localized PCa. Our findings have led to a change in our department’s diagnostic strategy implementing a policy of offering pre-biopsy MRI scans for all biopsy-naïve men who are candidates for curative treatment if diagnosed with sPCa. Based on patient-shared decision making then men are stratified by bpMRI suspicion score and PSAd where immediate biopsies are only recommended for men with either suspicious bpMRI results (score ≥4), PSAd ≥0.15 ng/mL/cc or additional high-risk factors such as family history of PCa, a high score on a risk-calculator or known germline mutations (e.g., BRCA2).
Written by: Lars Boesen, Nis Nørgaard, Vibeke Løgager, Ingegerd Balslev, Rasmus Bisbjerg, Karen-Cecilie Thestrup, Henrik Jakobsen, Henrik S Thomsen
Department of Urology, Herlev Gentofte University Hospital, Herlev, Denmark. Electronic address: ., Department of Urology, Herlev Gentofte University Hospital, Herlev, Denmark., Department of Radiology, Herlev Gentofte University Hospital, Herlev, Denmark., Department of Pathology, Herlev Gentofte University Hospital, Herlev, Denmark.
Read the Abstract
Figure 1. Prostate cancer detection rates for all patients (N = 808) based on combined biopsy results stratified by bpMRI suspicion scores, PSA density, and PCa significance.
Overall, we concluded that a simple and rapid bpMRI together with PSAd may be used as a combined triage test to safely avoid prostate biopsies in biopsy-naïve men with clinical suspicion of localized PCa. Our findings have led to a change in our department’s diagnostic strategy implementing a policy of offering pre-biopsy MRI scans for all biopsy-naïve men who are candidates for curative treatment if diagnosed with sPCa. Based on patient-shared decision making then men are stratified by bpMRI suspicion score and PSAd where immediate biopsies are only recommended for men with either suspicious bpMRI results (score ≥4), PSAd ≥0.15 ng/mL/cc or additional high-risk factors such as family history of PCa, a high score on a risk-calculator or known germline mutations (e.g., BRCA2).
Written by: Lars Boesen, Nis Nørgaard, Vibeke Løgager, Ingegerd Balslev, Rasmus Bisbjerg, Karen-Cecilie Thestrup, Henrik Jakobsen, Henrik S Thomsen
Department of Urology, Herlev Gentofte University Hospital, Herlev, Denmark. Electronic address: ., Department of Urology, Herlev Gentofte University Hospital, Herlev, Denmark., Department of Radiology, Herlev Gentofte University Hospital, Herlev, Denmark., Department of Pathology, Herlev Gentofte University Hospital, Herlev, Denmark.
Read the Abstract