The use of 29 MHz transrectal micro-ultrasound to stratify the prostate cancer risk in patients with PI-RADS III lesions at multiparametric MRI: A single institutional analysis.

INTRODUCTION

Magnetic Resonance Imaging (MRI) has emerged as the most accurate diagnostic tool, showing a high sensitivity in the diagnosis of clinically significant prostate cancer (csCaP). However only a minority of patients with a PI-RADS 3 lesion at multiparametric magnetic resonance imaging (MRI) are diagnosed with csCaP. The aim of the current study was to assess whether high resolution micro-ultrasound (microUS) could help in sub-stratifying the risk of csCaP in this specific population.

MATERIAL AND METHODS

We retrospectively analyzed the records of 111 consecutive patients scheduled for a prostate biopsy with at least 1 PI-RADS 3 lesions at MRI. We excluded patients with a PIRADS >3 lesion, even if they had a coexisting PIRADS 3 lesions. MicroUS was performed in all patients before prostate biopsy by an operator blind to MRI results. The Prostate Risk Identification using MicroUS (PRI-MUS) protocol was used to assess the risk of CaP and csCaP. All patients received both targeted and systematic biopsies. The primary endpoint was to determine the diagnostic accuracy of microUS in detection of csCaP in patients with a PI-RADS 3 lesion at MRI. Specifically, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of microUS were determined. Multivariable logistic regression models (MLRMs) were fitted to identify predictors of CaP. The diagnostic accuracy was reported as area under the receiver operator characteristic (ROC) curve.

RESULTS

Overall, 43 patients (38.7%) harboured CaP and 22 (20%) csCaP. MicroUS showed a high sensitivity and negative predictive value (100%), while its specificity and positive predictive value were 33.7% and 27.2%, respectively. Among patients without lesions at microUS, 25 (83.3%) did not harbour CaP, while 5 (16.7%) patients were diagnosed with a Gleason score 6 CaP, with no patients harbouring csCaP. Using microUS, the csCaP detection would have remained 100%, while reducing the detection of insignificant CaP of a 23.8% extent (n = 5). In MLRMs, lesion identified at microUS and continuously-coded PSAd were independent predictors of CaP. The accuracy of a model including PRI-MUS score, digital rectal examination (DRE), PSA density, age and family history was 0.744 (95% CI: 0.645 - 0.843).

CONCLUSION

In our single-institutional retrospective study, microUS was potentially capable to stratify the presence of CaP in patients with an equivocal MRI. Further prospective studies on larger populations are needed to validate our results.