MP56-16 A NEW SCORING SYSTEM USING PROSTATE IMAGING REPORTING AND DATA SYSTEM (PI-RADS) TO DETECT SIGNIFICANT CANCER FOR MAGNETIC RESONANCE IMAGING-TARGETED PROSTATE BIOPSY

Abstract

INTRODUCTION AND OBJECTIVE: Prostate Imaging Reporting and Data system (PI-RADs) is associated with significant cancer (SC) detection. Although PI-RADs≥ 3 is considered to be positive, there are many reports suggesting the significant different of the detection rates of SC between PI-RADs 3 and 4-5. In using PIRADs for MRI-targeted prostate biopsy (TgB), what kind of men have SC have been not fully examined. In this study, we examined the detective ability between PI-RADs 3/4-5 and developed a new scoring system for prediction for SC detection of TgB using PI-RADs. METHODS: Between April 2013 and April 2019, 470 men with PSA <20 ng/mL received TgB. PI-RADs ver. 2 was used and considered as positive in ≥3. TgB was performed 4 cores for a lesion and TgB was performed for each lesions with multiple positive PI-RADs. SC in this study was defined as Gleason Score ≥ 3+4 or maximum cancer length ≥ 5 mm. We assessed the association between the detection of SC and PI-RADs using multivariable regression analyses. Clinical variables examined included age, PSA, DRE finding, prostate volume (PV), TgB location (TZ/PZ), and PI-RADS score (3/4-5). We developed a new scoring system with first 300 cases using estimated value of multivariable analysis; we validated it with the latter 170 cases. Predictive accuracy and performance characteristics were assessed using the area under the receiver operating characteristic curve (AUC), respectively. RESULTS: In first 300 cases, the median age and PSA were 69 years and 7.7 ng/mL, respectively. Location of TgB was TZ/PZ in 75/225 men. PI-RADs were 3/4/5 in 65/178/57 men. SC was detected in 160 (53 %) men of TgB. In multivariable analysis, PI-RADS 4-5 (odds ratio [OR] 5.9, p<.01), the positive DRE (OR 4.4, p<.01), lower PV (< 30 mL, OR 2.6, p<.01), higher PSA (> 8.0 ng/mL, OR 2.4, p<.01) and PZ (OR 1.9, p< .01) were significantly associated with positive TgB for SC. From the results of the analysis, we assigned 1 to 3 points to each variable and developed a scoring system predicting positive TgB for SC (total of 0-8 points, as shown in Table) and AUC for detecting SC was 0.813. In validated group (170 cases) for this new scoring system, AUC for detecting SC was 0.811. CONCLUSIONS: We developed a new scoring system using PI-RADs for detecting SC by TgB and verified its validity. PIRDAS3 might not predict SC.Source of Funding: none