Prostate Cancer Genetic Risk Calculator 1.0

Joseph C. Presti, Jr., MD, FACS, Regional Leader of Urologic Oncology Surgery at Kaiser Permanente Northern California, discusses the development of Kaiser Permanente’s new prostate cancer risk calculator and its merits. Dr. Presti explains that older risk calculators tend to oversimplify variables like race, are based on outdated systematic biopsy schemes, and are often poorly calibrated due to the sampling frame used. Using TRIPOD guidelines (transparent reporting of a multivariable prediction model for individual prognosis or diagnosis) and the LASSO (least absolute shrinkage and selection operator) system of selection, Kaiser Permanente researchers determined that the variables that ought to be included in a prostate cancer prediction model are age, race, PSA, body-mass index, family history, number of prior negative biopsies, digital rectal exam (DRE), and prostate volume. They created 3 different models based on this, with the simplest but least accurate including clinical core variables but no DRE and no prostate volume, the second most accurate including DRE but no prostate volume, and the most accurate including DRE and prostate volume. Dr. Presti notes that all of these models compare favorably to other risk calculators.