The FUTURE randomised trial showed no difference in prostate cancer (PCa) detection between cognitive, in-bore and software guided MRI-targeted biopsy (TBx) techniques, a finding supported by meta-analyses and reflected in European Association of Urology (EAU) guidelines. However, whether any TBx method better predicts final pathology at radical prostatectomy (RP) has remained unclear. This study compared cognitive versus software-guided TBx using RP histopathology as reference. From the PROMOD international database, 9966 patients were screened, and 1045 met inclusion criteria: positive MRI (PI-RADS>2), PCa detected on combined systematic biopsy (SBx) and TBx, and subsequent RP between 2015 and 2022. Exclusions included prior prostate surgery, prostate specific antigen (PSA) ≥50ng/mL, in-bore or mixed TBx techniques, and incomplete MRI data. Of the cohort, 31% underwent cognitive TBx and 69% software-guided TBx. The primary outcome was concordance between TBx and RP International Society of Urological Pathology (ISUP) Grade Group (GG), with upgrading and downgrading also assessed. Statistical analyses included chi-squared testing, multivariable logistic regression, and 1:1 propensity-score matching. For combined SBx + TBx, concordance was 60% for software-guided TBx versus 52% for cognitive TBx; upgrading was higher with cognitive TBx. Using TBx cores alone, both approaches showed similar accuracy. Propensity-score matching produced a well-balanced cohort of 540 patients, again 6 demonstrating no significant difference in concordance between techniques. Adding SBx consistently improved accuracy and reduced upgrading, regardless of fusion method. Multivariable analysis confirmed no association between fusion technique and concordance, although the transrectal route was linked to lower accuracy. Despite similar overall concordance, cognitive TBx showed higher rates of GG upstaging and missed more clinically significant PCa. Software-guided TBx demonstrated more favourable risk-group stability from biopsy to RP. Limitations include the retrospective design, lack of central MRI review, and missing operator-related data. In conclusion, neither fusion technique outperformed the other in predicting final ISUP GG, aligning with current EAU guidance. Nonetheless, software-guided TBx may provide more reliable preoperative risk assessment, while SBx remains important for optimal grade prediction before RP.
Accuracy of cognitive vs software-guided MRI-targeted biopsy
Reviewed by Asif H Ansari
Accuracy of cognitive vs software-guided MRI-targeted biopsy in predicting final grading at prostatectomy.
