Abstract
BACKGROUND AND OBJECTIVE: Magnetic resonance imaging (MRI)-targeted biopsies for prostate cancer diagnosis can be performed as a transperineal biopsy (TP-Bx), yet its comparative performance with transrectal biopsy (TR-Bx) remains uncertain. We evaluated the detection of clinically significant prostate cancer (csPCa) by TP-Bx and TR-Bx according to lesion size and location using Prostate Imaging Reporting & Data System (PI-RADS) criteria. METHODS: We retrospectively reviewed biopsy-naïve patients who underwent MRI-guided prostate biopsies at our institution. TR-Bx was performed using MRI-ultrasound fusion in the urology clinic, while TP-Bx utilized an MRI in-bore technique in the radiology department. Lesions were stratified by PI-RADS-defined score, location, and size. RESULTS: Among 200 patients (100 TP-Bx, 100 TR-Bx), 276 PI-RADS score 3 to 5 lesions were biopsied (141 TP-Bx, 135 TR-Bx). Overall csPCa detection did not differ significantly between TP-Bx and TR-Bx (28% vs. 38%; OR = 1.5, CI = 0.9-2.6, P = 0.12). However, TR-Bx detected significantly more csPCa in non-apical peripheral zone (PZ) lesions compared to TP-Bx [45% (36/80) vs. 29% (21/73), adjusted OR = 4.6, 95% CI = 1.29-16.4, P = 0.019], particularly for small (diameter ≤1 cm) lesions (35% [16/46] vs. 12% [5/42], adjusted OR = 8.06, 95% CI = 1.45-44.7, P = 0.017). No significant difference was observed for larger lesions (diameter >1 cm). CONCLUSIONS: Overall csPCa detection rates were comparable between TP-Bx and TR-Bx, with no statistically significant difference. However, TR-Bx demonstrated superior detection in small non-apical PZ lesions, suggesting an anatomic and size-dependent advantage. These exploratory findings support further prospective studies to refine MRI-targeted biopsy protocols using PI-RADS-defined lesion characteristics to inform personalized biopsy strategies.