An Ai Driven Approach To Predict The Outcome Of Prostate Biopsy: Identifying Cancer, Clinically Significant Disease, And Unfavorable Pathological Features On Prostate Biopsy

You have accessJournal of UrologyProstate Cancer: Detection & Screening III (MP30)1 Sep 2021MP30-02 AN AI DRIVEN APPROACH TO PREDICT THE OUTCOME OF PROSTATE BIOPSY: IDENTIFYING CANCER, CLINICALLY SIGNIFICANT DISEASE, AND UNFAVORABLE PATHOLOGICAL FEATURES ON PROSTATE BIOPSY Cristina Pasat-Karasik, Lily Davenport, John Pfail, Marla Gabriele, Mae Gerenia, Kacie Schlussel, Ian Haas, Roy Berryhill, Samia Choudhury, Parita Ratnani, Bhavya Shukla, Vinayak Wagaskar, Peter Wiklund, Dara Lundon, and Ashutosh Tewari Cristina Pasat-KarasikCristina Pasat-Karasik More articles by this author , Lily DavenportLily Davenport More articles by this author , John PfailJohn Pfail More articles by this author , Marla GabrieleMarla Gabriele More articles by this author , Mae GereniaMae Gerenia More articles by this author , Kacie SchlusselKacie Schlussel More articles by this author , Ian HaasIan Haas More articles by this author , Roy BerryhillRoy Berryhill More articles by this author , Samia ChoudhurySamia Choudhury More articles by this author , Parita RatnaniParita Ratnani More articles by this author , Bhavya ShuklaBhavya Shukla More articles by this author , Vinayak WagaskarVinayak Wagaskar More articles by this author , Peter WiklundPeter Wiklund More articles by this author , Dara LundonDara Lundon More articles by this author , and Ashutosh TewariAshutosh Tewari More articles by this author View All Author Informationhttps://doi.org/10.1097/JU.0000000000002027.02AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract INTRODUCTION AND OBJECTIVE: Reducing unnecessary prostate biopsies is a crucial step towards reducing overdiagnosis. As we move toward more personalized medicine and individualized medical decision making, there is a fundamental need for better risk assessment tools that can aid patients and physicians in this decision-making process. The objective of this work was to construct a risk prediction model for patients suitable for prostate biopsy; to predict the presence of prostate cancer, clinically significant cancer (GGG ≥2), and unfavorable pathology (pT3a or pT3b and Gleason Grade Group≥3) on initial biopsy. METHODS: A prospectively maintained database of patients at risk for prostate cancer and undergoing prostate biopsy between January 2014 to December 2019, was queried. Descriptive statistics including frequencies and proportions were reported for categorical variables, while medians and interquartile ranges (IQRs) were reported for continuous variables. Data was divided into a training and holdout testing set using an 80%:20% ratio. The training set was used to train the classifiers and the hold-out set was used to evaluate the predictive ability of the model. Models were assessed for discriminative ability, calibration and clinical utility RESULTS: Study cohort consisted of 2,734 patients who underwent systemic TRUS biopsy for evaluation of prostate cancer. Overall, 1,528 (55.9%) patients had prostate cancer, 1,318 (48.2%) had clinically significant prostate cancer, and 725 (26.5%) had unfavorable pathology on final biopsy report. The AUCs of the models I the hold-out dataset were 82% (95% CI: 80.4-83.5), 83.8% (95% CI: 82.3-85.2), and 88.1% (95% CI: 86.7-89.5) for predicting prostate cancer, clinically significant disease, and unfavorable pathology, respectively, and outperformed other widely used risk-tools, in this cohort. CONCLUSIONS: Such multivariable risk prediction models can be used to personalize care and further aid in patient counseling for those undergoing prostate biopsy. Source of Funding: N/A © 2021 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 206Issue Supplement 3September 2021Page: e499-e499 Advertisement Copyright & Permissions© 2021 by American Urological Association Education and Research, Inc.MetricsAuthor Information Cristina Pasat-Karasik More articles by this author Lily Davenport More articles by this author John Pfail More articles by this author Marla Gabriele More articles by this author Mae Gerenia More articles by this author Kacie Schlussel More articles by this author Ian Haas More articles by this author Roy Berryhill More articles by this author Samia Choudhury More articles by this author Parita Ratnani More articles by this author Bhavya Shukla More articles by this author Vinayak Wagaskar More articles by this author Peter Wiklund More articles by this author Dara Lundon More articles by this author Ashutosh Tewari More articles by this author Expand All Advertisement Loading ...