Rajan T Gupta1,2, Alison Flanagan Brown1, Rachel Kloss Silverman3, Kae Jack Tay4, John F Madden5, Daniel J George2,6, Thomas J Polascik2,4. 1. 1 Department of Radiology, Duke University Medical Center, DUMC Box 3808, Durham, NC 27710. 2. 2 Duke Cancer Institute, Duke University Medical Center, Durham, NC. 3. 3 Department of Biostatistics, Gillings School of Global Public Health, The University of North Carolina at Chapel Hill, Chapel Hill, NC. 4. 4 Department of Surgery, Division of Urologic Surgery and Duke Prostate Center, Duke University Medical Center, Durham, NC. 5. 5 Department of Pathology, Duke University Medical Center, Durham, NC. 6. 6 Department of Medicine, Division of Medical Oncology, Duke University Medical Center, Durham, NC.
Abstract
OBJECTIVE: The purpose of this study is to investigate the accuracy of multiparametric MRI with endorectal coil and Partin tables in predicting organ-confined (OC) prostate cancer in a contemporary cohort undergoing radical prostatectomy (RP) and to assess the possible added value of radiologic staging based on multiparametric MRI to the predictive accuracy of Partin tables. MATERIALS AND METHODS: One hundred fifty-eight consecutive subjects underwent 3-T multiparametric MRI with endorectal coil before RP between November 2010 and November 2013. Data were randomly split 60% and 40% into derivation (n = 95) and validation (n = 62) datasets. Multiparametric MRI was used to assess the radiologic stage, and logistic regression models were created using the derivation dataset and were fit on the independent validation dataset using multiparametric MRI staging alone and with prostate-specific antigen (PSA) level as the covariate. The probability of each patient to harbor OC disease was calculated using an updated version of Partin tables, using either clinical staging from digital rectal examination (DRE) or radiologic staging (multiparametric MRI). The AUC was calculated to evaluate accuracy of these predictive methods. RESULTS: The accuracy of multiparametric MRI to predict OC disease on pathologic analysis was greater (AUC, 0.88) than that of Partin tables (AUC, 0.70) and improved when multiparametric MRI was combined with PSA level (AUC, 0.91). The accuracy of Partin nomograms to predict OC disease decreased (AUC, 0.63) when staging was based on multiparametric MRI versus DRE. CONCLUSION: The superior predictive accuracy of multiparametric MRI compared with Partin tables to predict OC disease validates the results of smaller previously published studies. Although there is no added benefit of substituting multiparametric MRI stage for clinical stage when using Partin tables, multiparametric MRI staging information is valuable as a stand-alone test.
OBJECTIVE: The purpose of this study is to investigate the accuracy of multiparametric MRI with endorectal coil and Partin tables in predicting organ-confined (OC) prostate cancer in a contemporary cohort undergoing radical prostatectomy (RP) and to assess the possible added value of radiologic staging based on multiparametric MRI to the predictive accuracy of Partin tables. MATERIALS AND METHODS: One hundred fifty-eight consecutive subjects underwent 3-T multiparametric MRI with endorectal coil before RP between November 2010 and November 2013. Data were randomly split 60% and 40% into derivation (n = 95) and validation (n = 62) datasets. Multiparametric MRI was used to assess the radiologic stage, and logistic regression models were created using the derivation dataset and were fit on the independent validation dataset using multiparametric MRI staging alone and with prostate-specific antigen (PSA) level as the covariate. The probability of each patient to harbor OC disease was calculated using an updated version of Partin tables, using either clinical staging from digital rectal examination (DRE) or radiologic staging (multiparametric MRI). The AUC was calculated to evaluate accuracy of these predictive methods. RESULTS: The accuracy of multiparametric MRI to predict OC disease on pathologic analysis was greater (AUC, 0.88) than that of Partin tables (AUC, 0.70) and improved when multiparametric MRI was combined with PSA level (AUC, 0.91). The accuracy of Partin nomograms to predict OC disease decreased (AUC, 0.63) when staging was based on multiparametric MRI versus DRE. CONCLUSION: The superior predictive accuracy of multiparametric MRI compared with Partin tables to predict OC disease validates the results of smaller previously published studies. Although there is no added benefit of substituting multiparametric MRI stage for clinical stage when using Partin tables, multiparametric MRI staging information is valuable as a stand-alone test.
Entities:
Keywords:
Partin tables; multiparametric prostate MRI; predictive nomograms; prostate cancer
Authors: Rossano Girometti; Martina Pancot; Marco Andrea Signor; Martina Urbani; Luca Balestreri; Chiara Zuiani Journal: Radiol Med Date: 2018-05-12 Impact factor: 3.469
Authors: Ariel A Schulman; Christina Sze; Efrat Tsivian; Rajan T Gupta; Judd W Moul; Thomas J Polascik Journal: Curr Urol Rep Date: 2017-07 Impact factor: 3.092
Authors: Maria A Gosein; Dylan Narinesingh; Shastri Motilal; Adrian P Ramkissoon; Cristal M Goetz; Kristy Sadho; Murrie D Mosodeen; Renee Banfield Journal: Radiol Imaging Cancer Date: 2020-07-31
Authors: Geoffrey S Gaunay; Vinay Patel; Paras Shah; Daniel Moreira; Ardeshir R Rastinehad; Eran Ben-Levi; Robert Villani; Manish A Vira Journal: Asian J Urol Date: 2016-11-19