For this study, the authors identified 135 patients, from a prospectively maintained prostate biopsy database, who had a negative mpMRI (3T; DWI, T2W, DCE) at a single institution prior to systematic 12-core prostate biopsy. Negative mpMRI was defined as a PIRADS score reported by an experienced radiologist. Patients with mpMRI done elsewhere or with any prior treatment for prostate cancer were excluded. Clinically significant prostate cancer was defined as Gleason > 3+4 on prostate biopsy. PSA density was defined as PSA/prostate volume on MRI. Pearson’s chi-square or Fisher’s exact tests, or Kruskal-Wallis, univariate and multivariable logistic regression were performed for statistical analyses (significant if p<0.05).
The median (IQR) age was 64 years (58-69), PSA was 5.9 ng/mL (4.1-8.0), prostate volume was 55cc (38-79), and PSAD was 0.1 ng/mL/cc (0.07-0.15). The patients were prostate biopsy-naïve in 36% (n=48), 44% (n=60) had previous negative prostate biopsy, and 20% (n=27) had previous positive prostate biopsy. The prostate cancer and clinically significant prostate cancer detection rate was 38% and 17%, respectively, corresponding to a NPV of 62% and 83%, respectively. Clinically significant prostate cancer detection rate was 21% for prostate biopsy-naïve, 8.3% for previous negative prostate biopsy and 33% for previous positive prostate biopsy patients. Patients with clinically significant prostate cancer (n=24) had smaller prostate volume (p=0.00036) and higher PSA density (p=0.000037), when compared to non-clinically significant prostate cancer patients (n=111). Multivariable analyses showed PSA density <0.15ng/mL/cc and previous negative prostate biopsy as independent predictors for no clinically significant prostate cancer on prostate biopsy (p=0.0001; p=0.0051). The NPV of mpMRI improves accordingly to different PSA density cutoffs. The PSA density <0.15ng/mL/cc <0.10ng/mL/cc and <0.07ng/mL/cc, correspond to NPV of 90%, 94% and 96% for clinically significant prostate cancer, respectively. For patients with previous negative prostate biopsy, a negative mpMRI combined with low PSA density further improves the NPV of the mpMRI. All patients with PSA density < 0.15ng/mL/cc and previous negative previous biopsy had no clinically significant prostate cancer on previous biopsy, yielding NPV of 100%.
Based on these results, the authors concluded that the negative predictive value of mpMRI is 83% for clinically significant prostate cancer. In case of negative mpMRI, patients with low PSA density, especially those with history of prior negative biopsy, may safely avoid prostate biopsy. Conversely, patients with high PSA density should consider prostate biopsy even in the setting of negative mpMRI.
Presented by: Giovanni E. Cacciamani, University of Southern California, USC Institute of Urology, Los Angeles, CA
Co-Authors: Oishi M, Ashrafi A, Shin T, Ohe C, Ghodoussipour S, Lin-Brande M, Winter M, Medina L, Margaryan T, Palmer S, Aron M, Ukimura O, Gill I, De Castro Abreu AL
Written by: Zachary Klaassen, MD, Urologic Oncology Fellow, University of Toronto, Princess Margaret Cancer Centre, twitter: @zklaassen_md at the 2018 European Association of Urology Meeting EAU18, 16-20 March, 2018 Copenhagen, Denmark