Comparison of prostate biopsy with or without pre-biopsy multi-parametric MRI in prostate cancer detection: an observational cohort study.
Bryant RJ., Hobbs CP., Eyre KS., Davies LC., Sullivan ME., Shields W., Sooriakumaran P., Verrill CL., Gleeson F., MacPherson R., Hamdy FC., Brewster SF.
PURPOSE: We hypothesized that 1) introducing pre-biopsy multiparametric magnetic resonance imaging (PB-mpMRI) increases the diagnostic yield of transrectal prostate biopsy (TRPB), and 2) this would inform recommendations regarding systematic TRPB in the "negative" PB-mpMRI setting. MATERIALS AND METHODS: Nine hundred and ninety-seven biopsy-naïve patients underwent TRPB alone to June 2016 (cohort A), and seven hundred and ninety-two underwent TRPB following PB-mpMRI thereafter (cohort B). Patients with PB-mpMRI lesions underwent "cognitive-targeted" plus systematic TRPB. Patients without lesions underwent systematic TRPB. RESULTS: Cohort B comprised younger (68 v 69 years, p=0.01) men with lower PSA (7.6 v 7.9 ng/mL, p=0.024) and prostate volume (56.1 v 62cc, p=0.006). There was no increase in overall PCa detection (57.6 v 56.7%, p=0.701), Gleason grade group (GGG), or number of positive cores (p>0.05 for each) in Cohort B versus Cohort A. Increased multi-focal prostatic intraepithelial neoplasia, maximum PCa core length (≥5mm versus <5mm), and radical surgery/HIFU (p<0.05 for each) was observed in Cohort B. Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of a "negative" PB-mpMRI for GGG 2-5 PCa were 88.1%, 59.8%, 67.8% and 84% respectively. For "negative" PB-mpMRIs, a PSAd≥0.15 cut-point only increased "clinically significant" PCa detection if the latter was defined as GGG 3-5 disease and/or tumor length ≥6mm. CONCLUSIONS: Introducing PB-mpMRI in our clinical setting increased the diagnostic yield of PCa per biopsy core. Not performing a systematic TRPB when the PB-mpMRI was "negative" would have led to under-detection of 15.1% (approximately 1 in 6) of GGG≥2 PCa cases.