Abstract / Summary
Background/Objectives: Multiparametric magnetic resonance imaging (mpMRI)-targeted biopsy has become central to prostate cancer (PCa) diagnosis. While cognitive and software-based fusion targeting have already been compared, the use of high-resolution micro-ultrasound as a real-time guidance platform for cognitive translation of MRI-defined lesions remains incompletely characterized. This study prospectively compared mpMRI-transrectal ultrasound (TRUS) fusion-guided transperineal (TP) biopsy with cognitive MRI-targeted TP biopsy under real-time ExactVu® micro-ultrasound guidance. Methods: We conducted a prospective, observational, comparative study including 58 consecutive patients with suspected PCa and at least one lesion on pre-biopsy mpMRI with a Prostate Imaging Reporting and Data System (PI-RADS) score ≥ 3. Twenty-nine patients were included in each group. The analysis used a standardized 16-core TP biopsy set comprising four targeted cores from the MRI-defined index lesion and 12 systematic cores. The primary endpoint was clinically significant PCa (csPCa) detection, defined as International Society of Urological Pathology (ISUP) Grade Group (GG) ≥ 2. Results: Overall, PCa detection was 65.5% in the fusion-guided biopsy group and 55.2% in the cognitive MRI-targeted biopsy group (absolute risk difference, 10.3 percentage points; 95% CI, −14.1 to 33.2; p = 0.592). csPCa detection was identical in the two groups (37.9% each; absolute risk difference, 0.0 percentage points; 95% CI, −23.6 to 23.6; p = 1.000). In exploratory multivariable logistic regression, biopsy technique was not associated with csPCa detection after adjustment (adjusted OR 0.73, 95% CI 0.22–2.40; p = 0.605), whereas prostate-specific antigen density (PSAD) showed an association with csPCa detection in the exploratory multivariable model (adjusted OR 1.96 per 0.1-unit increase, 95% CI 1.04–3.71; p = 0.039). Conclusions: Cognitive MRI targeting under real-time micro-ultrasound guidance showed the same observed csPCa detection rate as software fusion within a standardized targeted-plus-systematic biopsy pathway. Wide confidence intervals preclude conclusions regarding equivalence or non-inferiority.