Choosing Number of Cores in a Prostate Cancer Biopsy Session to Achieve Desired Specificity-Sensitivity Trade-Off

• Activity Number: 615
• Type: Contributed
• Date/Time: Thursday, August 2, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Gerald Ogola*+ and Robert Serfling
• Companies: Baylor Health Care System and The University of Texas at Dallas
• Keywords: Prostate cancer biopsy ; Sensitivity-specificity trade-off ; Model-based nomogram

Abstract

Planning effective prostate cancer biopsy sessions challenges urologists. One must choose the number of biopsy cores to minimize the burden on patients while also achieving high probability of detecting significant cancer if present. In this work, the probability distribution of the number of distinct cores hit in a session is derived as a function of prostate volume, tumor volume, and number of cores. This yields specificity and sensitivity values for any given criterion for deciding "insignificant" versus "significant" cancer based on the number of cores hit. A nomogram is constructed for choosing the number of cores for a prostate of given volume so as to obtain the best trade-off between specificity and sensitivity. Such decision criteria improve upon existing ones for evaluation of biopsy results and provide a new tool for addressing the dilemma of avoiding overtreatment of prostate cancer. Bayesian extensions using prior distributions on tumor volume are also discussed.