MP20-12 PREDICTIVE VALUE OF TUMOR CONTACT LENGTH ON MAGNETIC RESONANCE IMAGING FOR EXTRACAPSULAR EXTENSION OF PROSTATE CANCER

Abstract

You have accessJournal of UrologyProstate Cancer: Staging I1 Apr 2017MP20-12 PREDICTIVE VALUE OF TUMOR CONTACT LENGTH ON MAGNETIC RESONANCE IMAGING FOR EXTRACAPSULAR EXTENSION OF PROSTATE CANCER Kazumi Kamoi, Koji Okihara, Fumiya Hongo, Yasuyuki Naitoh, Atsuko Iwata, Motohiro Kanazawa, So Ushijima, and Osamu Ukimura Kazumi KamoiKazumi Kamoi More articles by this author , Koji OkiharaKoji Okihara More articles by this author , Fumiya HongoFumiya Hongo More articles by this author , Yasuyuki NaitohYasuyuki Naitoh More articles by this author , Atsuko IwataAtsuko Iwata More articles by this author , Motohiro KanazawaMotohiro Kanazawa More articles by this author , So UshijimaSo Ushijima More articles by this author , and Osamu UkimuraOsamu Ukimura More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2017.02.644AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES Tumor contact length (TCL) is defined as the amount of prostate cancer in contact with the prostatic capsule. We evaluated the ability of magnetic resonance imaging (MRI) determined tumor contact length to predict pathological extracapsular extension (pECE) compared to existing predictors of pECE. METHODS We retrospectively analyzed the records of 128 consecutive patients with MRI/ultrasound fusion targeted, biopsy proven prostate cancer who underwent robotic-assisted radical prostatectomy from April 2013 to July 2016. Median patient age was 67 years and median prostate specific antigen was 7.11 ng/ml. Clinical stage was cT1 in 71 cases (55%) and cT2 in 53 (41%). Postoperative pathological analysis confirmed pT2 in 90 patients (70%) and pT3 in 38 (30%). We evaluated 1) in the radical prostatectomy specimen the correlation of pECE with pathological cancer volume, pathological TCL and Gleason score, 2) the correlation between pECE and MRI determined TCL, and 3) the ability of preoperative variables to predict pECE. RESULTS Logistic regression analysis revealed that pathological TCL correlated better with pECE than the predictive power of pathological cancer volume (0.822 vs 0.659). The Spearman correlation between pathological and MRI determined TCL was r = 0.873 (p <0.0001). ROC AUC analysis revealed that m MRI determined TCL outperformed cancer core involvement on targeted biopsy and the Partin tables to predict pECE (0.84 vs 0.70 and 0.57, respectively). The best TCL thresholds for predicting pECE was 13.5 mm (sensitivity 71%, specificity 80%) and the predictability of pECE outperformed comparing with conventional MRI criteria (MRI-TCL: odds ratio of 10.0, p<0.0001, and MRI criteria: odds ratio of 1.15, p=0.8670). CONCLUSIONS MRI determined TCL could be a promising quantitative predictor of pECE. The best TCL threshold of 13.5mm might predict pECE with higher accuracy, although PI-RADSv2 reported tumor-capsule interface of greater than 1.0 cm as cut-off value for staging of pECE. © 2017FiguresReferencesRelatedDetails Volume 197Issue 4SApril 2017Page: e243 Advertisement Copyright & Permissions© 2017MetricsAuthor Information Kazumi Kamoi More articles by this author Koji Okihara More articles by this author Fumiya Hongo More articles by this author Yasuyuki Naitoh More articles by this author Atsuko Iwata More articles by this author Motohiro Kanazawa More articles by this author So Ushijima More articles by this author Osamu Ukimura More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...