Niu X-K, He W-F, Zhang Y, Das S K, Li J, Xiong Y, Wang Y-H
Department of Radiology, Affiliated Hospital of Chengdu University, Chengdu, 610081, China.
Department of Cardiology, Affiliated Hospital of North Sichuan Medical College, Sichuan 637000, China.
Clin Radiol. 2017 Jun;72(6):458-464. doi: 10.1016/j.crad.2016.12.005. Epub 2017 Jan 6.
To establish a predictive nomogram for high-grade prostate cancer (HGPCa) in biopsy-naive patients based on the Prostate Imaging-Reporting and Data System version 2 (PI-RADS v2), magnetic resonance imaging (MRI)-based prostate volume (PV), MRI-based PV-adjusted prostate-specific antigen density (PSAD), and other classical parameters.
Between August 2014 and August 2015, 158 men who were eligible for analysis were included as the training cohort. A prediction model for HGPCa was built using backward logistic regression and was presented on a nomogram. The prediction model was evaluated by a validation cohort between September 2015 and March 2016 (n=89). Histology of all lesions was obtained with MRI-directed transrectal ultrasound (TRUS)-guided targeted and sectoral biopsy.
The multivariate analysis revealed that patient age, PI-RADS v2 score, and adjusted PSAD were independent predictors for HGPCa. The most discriminative cut-off value for the logistic regression model was 0.33; the sensitivity, specificity, positive predictive value, and negative predictive value were 83.3%, 87.4%, 88.4%, and 81.2%, respectively. The diagnostic performance measures retained similar values in the validation cohort (AUC=0.83).
The nomogram for forecasting HGPCa is effective and potentially reducing harm from unnecessary prostate biopsy and over-diagnosis.
基于前列腺影像报告和数据系统第2版（PI-RADS v2）、磁共振成像（MRI）测定的前列腺体积（PV）、基于MRI的PV校正前列腺特异性抗原密度（PSAD）以及其他经典参数，为未经活检的患者建立高级别前列腺癌（HGPCa）预测列线图。
2014年8月至2015年8月期间，158名符合分析条件的男性被纳入训练队列。使用向后逻辑回归建立HGPCa预测模型，并在列线图上展示。2015年9月至2016年3月期间的验证队列（n = 89）对预测模型进行评估。所有病变的组织学检查通过MRI引导的经直肠超声（TRUS）引导下的靶向和扇形活检获得。
多变量分析显示患者年龄、PI-RADS v2评分和校正后的PSAD是HGPCa的独立预测因素。逻辑回归模型最具判别力的截断值为0.33；敏感性、特异性、阳性预测值和阴性预测值分别为83.3%、87.4%、88.4%和81.2%。验证队列中的诊断性能指标保持相似值（AUC = 0.83）。
预测HGPCa的列线图有效，可能减少不必要的前列腺活检和过度诊断带来的危害。
