Comparative Impact of Elastic Magnetic Resonance Imaging-Transrectal Ultrasound Fusion Biopsy on Concordance of Detection of Clinically Significant Prostate Cancer by International Society of Urological Pathology Grade in Biopsy-naïve Men with Prostate-specific Antigen ≤20 ng/ml and cT1-2 Disease.

Background and objective: Magnetic resonance imaging (MRI)-transrectal ultrasound (TRUS) fusion guided biopsy is the cornerstone of prostate cancer (PC) diagnosis. While prior studies have focused on detection rates, the impact of fusion registration methods-elastic registration techniques (ERT) versus rigid registration techniques (RRT)-on International Society of Urological Pathology (ISUP) grade concordance remains underexplored. Our objectives were to assess the effect of ERT versus RRT on the concordance between targeted biopsy (TBx) and overall biopsies (OBx) for detection of (1) clinically significant PC (csPC; defined as ISUP grade ≥2) and (2) high-grade PC (hgPC; defined as ISUP grade ≥3) in biopsy-naïve men with confirmed PC.

Methods: Our multicenter retrospective study included 888 biopsy-naïve men with confirmed PC (prostate-specific antigen ≤20 ng/ml, cT1-2, Prostate Imaging-Reporting and Data System [PI-RADS] score 3-5) who underwent MRI-TRUS fusion guided biopsy using ERT (n = 479) or RRT (n = 409) at two high-volume institutions. After 1:1 propensity score matching (PSM) to control for confounding, a sample of 674 patients was included in the final analysis. The primary endpoint was the concordance of csPC detection between TBx and OBx. Secondary endpoints included concordance for hgPC detection (ISUP grade ≥3) between TBx and OBx, concordance for hgPC (ISUP grade ≥3) between systematic biopsy (SBx) and OBx, biopsy sampling metrics, and subgroup analyses for PI-RADS 3 lesions. Multivariable logistic binomial regression models adjusted for clinical and imaging covariates were tested.

Key findings and limitations: There was a significant difference in the frequency of csPC concordance between the ERT and RRT groups (60.2% vs 33.6%; p < 0.0001). Moreover, the ERT approach was associated with significantly higher odds of being classified as concordant csPC in comparison to RRT (adjusted odds ratio [aOR] 4.82, 95% confidence interval 2.82-8.24). ERT was also significantly associated with higher odds of hgPC concordance after adjusting for PSA density, clinical TNM stage and PI-RADS score (aOR 2,51, 95% confidence interval 1.51-4.16; corrected p = 0.0014). ERT reduced overgrading of ISUP grade 1 lesions (12.5% vs 39.2%; p < 0.001). Despite lower core volume and fewer positive cores, ERT achieved similar maximum cancer core length, suggesting superior spatial targeting. PI-RADS 3 subgroup analyses showed favorable trends for ERT, although the results were not statistically significant (p >0.05).

Conclusions and clinical implications: ERT was associated with better concordance for detection of both csPC and hgPC on TBx, supporting more accurate risk stratification while reducing detection of indolent cancer. While our findings indicate diagnostic advantages for ERT over RRT, prospective multicenter studies with centralized pathology review are warranted for external validation and evaluation of the downstream clinical impact.