Diagnostic accuracy of multiparametric MRI- and microultrasound-targeted biopsy in biopsy-naïve patients with a PI-RADS 5 lesion: a single-institutional study.

ABSTRACT

PURPOSE: To evaluate the diagnostic accuracy of multiparametric magnetic resonance imaging (MRI)- and microultrasound (microUS)-guided targeted biopsy (TBx) in detecting prostate cancer (PCa) and clinically significant (cs) PCa among men with Prostate Imaging Reporting and Data System (PI-RADS 5) lesions and to compare this combined TBx (CTBx) strategy with CTBx plus systemic biopsy (SBx). METHODS: One hundred and thirty-six biopsy-naïve patients with PI-RADS 5 lesion at multiparametric MRI undergoing CTBx plus SBx were retrospectively evaluated. Analysis of diagnostic performance of microUS-TBx, MRI-TBx, CTBx, SBx and combined CTBx plus SBx was performed. Cost (downgrade, upgrade and biopsy core) to effectiveness (detection rate) was compared. RESULTS: CTBx achieved a comparable detection rate to CTBx plus SBx in diagnosis of PCa and csPCa (PCa 78.7% [107/136] vs 79.4% [108/136]; csPCa 67.6% [92/136] vs 67.6% [92/136]; p > 0.05) and outperformed SBx (PCa 58.8% [80/136]; csPCa 47.8% [65/136]; p < 0.001). Using CTB would have avoided 41.1% (56/136) unnecessary SBx, without missing any csPCa. The rate of any upgrading or csPCa upgrading was significantly higher by SBx than by CTBx [33/65 (50.8%) vs 17/65 (26.1%) and 20/65 (30.8%) vs 4/65 (6.15%), respectively, p < 0.05]. Considering csPCa detection rate, microUS showed high sensitivity and positive predictive value (94.6%, 87.9%, respectively), with lower specificity and negative predictive value (25.0% and 44.4%, respectively). At multivariable logistic regression models, positive microUS was identified as an independent predictor of csPCa (p = 0.024). CONCLUSIONS: A combined microUS/MRI-TBx approach could be the ideal imaging tool for characterizing primary disease in PI-RADS five patients, allowing SBx to be avoided.