2.1

Study Design and Patient Selection

We performed a retrospective database review of patients diagnosed with PCa who underwent robotic assisted RP in a single multi-site institution between January 2021 and June 2023, coinciding with integration of microUS at our site. Patients were included who had both prostate biopsy and final RP pathology available. Patients were excluded if there was no pathological information accessible about their prostate biopsy.

2.2

Procedure

Patients were placed in the left lateral decubitus position and had either a systematic transrectal conventional ultrasound guided biopsy [ ], MRI-TRUS fusion biopsy, microUS guided biopsy or MRI-microUS fusion biopsy. TRUS biopsy alone were patients who primarily received their TRUS biopsies at outside institutions (71%), and those with PIRADS≤2 findings on MRI. Patients received microUS biopsy without MRI fusion either because they went directly from elevated PSA to biopsy after shared decision making, had PIRADS <3 lesions on MRI, or were unable to have MRI because of claustrophobia or implants. Multiparametric MRI guidance utilized a 3T system and an external phased array. MRI uses a combination of multiplanar T2-weight imaging, diffusion weighted imaging and dynamic contrast enhanced imaging [ ]. A 1.5T system is used for patients with pacemakers or pelvic/hip prostheses. For MRI-TRUS fusion biopsy, the multiparametric MRI image is directly fused with the real-time US image using the Artemis ^TM system. A targeted biopsy of 3–5 cores per operator preference was performed for any clinically significant Prostate Imaging Reporting and Data System (PIRADS) lesion, followed by the standard 10–12 core template biopsy.

MicroUS biopsies were performed by 7 urologists using the ExactVu ^TM microUS imaging system. Suspicious lesions on the PRIMUS scale or by cognitive visualization were biopsied in 3–5 cores along with a standard templated biopsy. The cores for the standard templated biopsy were selected by the performing urologist and not the software system. The MRI-microUS fusion biopsy was performed using the Exact Imaging system software (Exact Imaging, Markham, Canada). Fusion biopsy choice, whether MRI-microUS fusion or MRI-TRUS fusion, was based on system availability and clinician preference.

RP was performed robotically, by 1 of 9 surgeons. RP approach, and the use of single port or multiport robot for prostatectomy was at the surgeon’s discretion.

2.3

Pathology Analysis

All prostate biopsies were independently reviewed by genitourinary pathologists and each biopsy core was assigned Gleason score based on the 2019 ISUP Gleason scoring system. For TRUS only biopsies, almost all (97.2%) pathology slides were read by intra-institutional pathologists regardless of whether their biopsy was performed within our institution or elsewhere. Similarly, RP specimens were received fresh, photographed, step sectioned and reviewed by two genitourinary pathologists for Gleason grading, specimen surgical parameters and cellular testing.

2.4

Study Outcomes

The primary outcome of the study was the occurrence of any Gleason Grade (GG) Group upgrade between biopsy and RP. The secondary outcomes included demographic and pathologic factors affecting upgrade, Gleason downgrading, defined as any decrease in pathologic GG from most recent biopsy, as well as GG concordance (equal Gleason grade group on biopsy and RP). In addition, major upgrade was defined as an increase in final pathology GG of ≥2 from GG at biopsy.

2.5

Statistical Analysis

We described clinical and demographic characteristics of the study cohort. For patients who received more than one biopsy, we considered findings on the biopsy in closest proximity to prostatectomy. We compared the distributions of patient clinical, demographic, and pathologic characteristics across biopsy types using Analysis of variance (ANOVA) and Pearson chi-squared analyses. Univariable and multivariable logistic regression were performed to identify factors associated with pathologic upgrade. Variables that were significant ( P < 0.05)on univariable analysis, and type of prostate biopsy, regardless of statistical significance, were included in the multivariate model. Pairwise comparisons were performed between biopsy techniques. P-values were reported as 2-sided and values less than 0.05 were considered statistically significant. Statistical analysis was performed using StataMP. This study was approved by the Yale University Institutional Review Board (IRB 2000037149)