Impact of MRI/US fusion‐guided prostate biopsy on biopsy‐naïve patients: A single urologist’s experience

Abstract Objectives To report our experience with imaging‐guided targeted prostate biopsy (IGTpBx) for patients undergoing initial prostate biopsy in a clinical setting. Materials and methods From July 2014 to February 2020, 305 men who had IGTpBx performed as their first prostate biopsy were enrolled. Two dedicated magnetic resonance imaging (MRI) radiologists segmented at least 1 region of interest (ROI) for each of these men using screening 1.5T MRI images. A single urologist employed the robotic‐assisted Artemis MRI/ultrasonography (US) fusion platform to obtain 2‐3 targeted samples from each ROI and additional random samples from the zones of the prostate outside the ROIs (a total of 12 zonal samples). Biopsy outcomes were categorized based on the Gleason score (GS) grade group (GG) as no cancer, favorable (GG < 3 or GS < 4 + 3), or clinically significant (GG ≥ 3 or GS ≥ 4 + 3) cancer. Results The overall cancer detection rate was 75%:31% clinically significant, 44% favorable, and 25% no cancer. These findings triggered active interventions in 176 (58%) patients. A prostate‐specific antigen (PSA) level of 0–4 ng/mL was detected in 39 (66%) of 59 patients (32 favorable, 7 significant), 4–10 ng/mL in 147 (77%) of 190 patients (85 favorable, 62 significant), and 10 ng/mL and over in 44 (80%) of 55 patients (17 favorable, 27 significant). Conclusions The tumor detection rate was 75% with IGTpBx in patients without a previous biopsy. In addition, about 42% of detected cancers were deemed clinically significant and led to active interventions. IGTpBx as a patient’s first prostate biopsy improves the detection of clinically significant prostate cancer when compared with historical data for random systematic prostate biopsy.


| INTRODUC TI ON
Prostate cancer (PCa) is one of the most common cancers in men worldwide. 1 Men with a high prostate-specific antigen (PSA) level and/or abnormal digital rectal examination (DRE) are often advised to undergo a prostate biopsy to establish or exclude the diagnosis.
Unlike other solid tumors, prostate cancer is detected using random biopsy sampling of the entire organ. The number of cores considered to be optimal for cancer detection has varied with time; the currently accepted technique is 10-to 12-core laterally directed transrectal ultrasound-guided biopsy (TRUS-bx). 2 This approach has inherent shortcomings: (1) a high false-negative detection rate of 20-24% and (2) an understaging rate of 50-80% for clinically significant prostate cancers. 3,4 Increasing the number of cores increases the detection of insignificant PCa. 5,6 In the United States, approximately 1.3 million prostate biopsies are performed annually 7 ; therefore, the overtreatment and undertreatment rates can have a significant effect, making an accurate diagnosis and avoidance of multiple unnecessary prostate biopsies increasingly important in PCa.
In the last decade, with the development of multiparametric magnetic resonance imaging (mpMRI), the ability to detect tumor has significantly improved. Many studies have shown that mpMRI could be used as a triage test to avoid unnecessary biopsy for insignificant cancer, such as with imaging-guided targeted prostate biopsy (IGTpBx). [8][9][10] IGTpBx is currently only recommended for selected patients with a prior negative biopsy and continued clinical suspicion of prostate cancer. 2 patients with no region of interest (ROI) biopsy samples obtained, and 6 patients with no random biopsies obtained (by physician decision), the final analysis included 305 patients ( Figure 1).
All 305 patients were selected after repeat review by one of two dedicated MRI radiologists who segmented at least 1 ROI as suspicious for cancer using the 5-point Likert system. This system assigns a score of 1, 2, 3, 4, or 5 to denote the probability of carcinoma as highly unlikely, unlikely, equivocal, likely, or highly likely, respectively. 12 The utilization of the Likert system scoring reflects our radiologists' practice. Although PI 13,14 Equivalent or better performance for clinically significant PCa detection has been reported with the Likert system, compared with PI-RADS, reducing the number of unnecessary prostate biopsies. 15,16 Although both systems are associated with inter-observer variability due to subjectivity, the Likert system may allow more freedom to assign a level of suspicion using all imaging sequences available. More importantly, the application of PI-RADS requires certain technical parameters for high-quality images (e.g., b-values of at least 1400 for DWI), which may not be achievable in some institutions (including ours at the time of the data collection). Outcomes were categorized as no cancer, favorable cancer (GG < 3 or GS < 4 + 3), or clinically significant cancer (GG ≥ 3 or GS ≥ 4 + 3). 17

| Statistical analysis
Statistical analysis was performed using the SPSS 24.0 software program for Windows (SPSS Inc., Chicago, IL, USA). Descriptive statistics for the clinical, pathological, and treatment-related data were provided.
Pearson's Chi-square test (or Fisher's exact test) was used to analyze categorical variables. Statistical significance was considered as P < .05.

| RE SULTS
Patient characteristics are provided in Table 1  MRI Likert score correlation with cancer detection also was in-  Table 4.

| D ISCUSS I ON
The main goal of prostate biopsy is to make an accurate diagnosis by detecting significant cancers while minimizing the false-negative rate and insignificant cancer detection. For many years, TRUS-bx has PCa detection, n (%) All In our study, 20 (9%) of 230 cancers were detected in only standard cores and 9 (45%) of 20 were GS ≥7 (2 of 20 were significant cancer).
The percentage of significant cancer that is detected in TRUS-bx but missed in IGTpBx in the literature varies from 0% to 10%. [22][23][24][25] In another study conducted by Maxeiner  and comprised 42% (96 of 230) of the detected cancers in our study ( Figure 1, Table 2). If we apply the criterion of GS ≥7 that is used in most studies, the significant cancer detection rate increases to 61% (187 of 305) and comprises 81% (187 of 230) of detected cancers ( Abbreviations: GG, Gleason group; GS, Gleason score; RARP, Robotassisted radical prostatectomy; ROI, region of interest.