Rather than simply asking whether cognitive and software fusion produce equivalent results, we compared three complete clinical pathways: in-office local-anesthetic (LA) freehand CF TPPB, general-anesthetic (GA) CF TPPB via brachytherapy grid, and GA SAF TPPB using the BiopSee platform. This distinction is important. Prior studies isolated the fusion modality; ours examined the entire diagnostic ecosystem — anesthetic setting, access technique, operator model, and institutional pathway — as it exists in real-world practice.
To account for the non-randomised, multicentre nature of the data, we used 1:1 nearest-neighbour propensity score matching with a caliper of 0.1, balancing groups on age, BMI, PSA density, prostate volume, PI-RADS score, and biopsy history. After matching, all standardised mean differences across covariates fell below 0.1, confirming adequate balance. Logistic regression incorporating matching weights was then used to estimate the odds of clinically significant prostate cancer (csPCa) detection. The results were robust across two prespecified sensitivity analyses — a multivariable regression in the full unmatched cohort and a PI-RADS-stratified matched analysis — both of which remained directionally consistent with the primary findings.
What emerged was striking: the LA freehand CF pathway was associated with significantly higher detection of csPCa than both the SAF pathway (OR 0.48, 95% CI 0.34–0.69; P<.001) and the GA grid-based CF pathway (OR 0.64, 95% CI 0.44–0.92; P=.017). The benefit was most pronounced in the PI-RADS 4–5 subgroup, where targeting precision arguably matters most.
The comparison with the brachytherapy grid pathway deserves particular attention. While grid-based access offers a structured approach that facilitates training, it introduces well-recognised anatomical constraints. Pubic arch interference can restrict needle trajectory, particularly in patients with larger prostates or prominent pubic arches, limiting access to the lateral peripheral zone and apical targets. Freehand transperineal access is not constrained in this way — it permits longer sampling vectors along the lateral peripheral zone and more flexible angulation toward the apex, areas where significant cancers are frequently undersampled by grid-based templates. These geometric differences in gland coverage may partly explain our findings and deserve dedicated prospective evaluation.
Software-assisted fusion carries its own limitations that are sometimes underappreciated in the literature. Registration error, intraprocedural prostate motion, and probe or patient movement can cause target drift between needle passes. Because re-registration after each core is generally impractical in routine practice, accumulated misalignment may reduce the accuracy of targeting — particularly for smaller or anteriorly located lesions.
Our findings do not invalidate software fusion or grid-based approaches. Both have important roles, particularly in centres building their transperineal programmes or training residents. Indeed, the inclusion of trainee operators in the Swiss cohort is a recognised limitation and may have contributed to the lower detection rates observed there. However, in the hands of experienced, high-volume operators — as was the case in both the New Zealand and Australian cohorts — freehand cognitive fusion under local anaesthesia performed at least as well or better than the technologically more intensive alternatives, while offering the additional advantages of avoiding general anaesthesia, reducing cost, and enabling office-based delivery of care.
Taken together, our data suggest that optimising the TPPB pathway is not simply a question of which fusion platform to adopt, but of how the entire procedure is organised and by whom. Future prospective multicentre trials with standardised MRI interpretation protocols, harmonised biopsy templates, controlled core numbers, and clearly defined operator experience thresholds are needed to disentangle the individual contributions of fusion modality, access technique, anesthetic setting, and operator expertise. Only then will we be able to confidently define which technical elements are truly driving differences in clinically significant prostate cancer detection.
Written by: Flavio Vasconcelos Ordones, Tauranga Public Hospital, Tauranga, Bay of Plenty, New Zealand; University of Auckland, Faculty of Medicine and Health Sciences, Auckland, New Zealand; Urology Department, UNESP, São Paulo State University, Botucatu, São Paulo, Brazil
References:
- Khoo et al. A Comparison of Prostate Cancer Detection between Visual Estimation (Cognitive Registration) and Image Fusion (Software Registration) Targeted Transperineal Prostate Biopsy. J Urol. 2021 Apr;205(4):1075-1081. doi: 10.1097/JU.0000000000001476.
- Hung et al. Same Target, Different Tools: A Multicenter Comparison of Clinically Significant Prostate Cancer Detection Across Three MRI-Targeted Transperineal Biopsy Techniques. JU Open Plus. June 2026. | DOI: 10.1097/JU9.0000000000000467.