Construction and Validation of a Diagnostic Prediction Model for Prostate Cancer in Patients With PI-RADS Score 3: a Multicenter, Retrospective Study

For patients with a PI-RADS score of 3, the diagnosis of prostate cancer is still use prostate biopsy, but the detection rates of prostate cancer and clinically significant prostate cancer are approximately 30% and 15%. It can be seen that most patients with PI-RADS 3 undergo unnecessary prostate biopsy and bear the risk of complications such as urinary tract infection. This makes most patients with PI-RADS 3 choose to refuse invasive prostate biopsy. Although researchers are committed to exploring biomarkers with high sensitivity and specificity, the application of biomarkers alone often cannot achieve the expected results. At present, the guidelines have recommended the use of diagnostic prediction models to assess patients' prostate cancer risk. Doctors and patients use diagnostic models to assess the risk of prostate cancer before prostate biopsy. For patients with a low probability of cancer, biopsy can be temporarily avoided, which to a certain extent reduces the phenomenon of prostate cancer overdiagnosis. This study plans to work with multiple medical centers to conduct statistical analysis based on existing prostate cancer screening markers combined with patients' clinical data such as prostate volume, prostate-specific antigen density, apparent diffusion coefficient, PI-RADS score and postoperative Gleason score, and then construct a prostate cancer diagnostic model to improve the diagnostic accuracy of prostate cancer for patients with PI-RADS score of 3. This will be of great significance for improving the early diagnosis of patients with PI-RADS 3 and reducing unnecessary prostate puncture biopsies.