Stojadinovic Miroslav, Jurisevic Nebojsa, Stojadinovic Milorad, Jankovic Slobodan
Faculty of Medical Sciences, University of Kragujevac, SvetozaraMarkovica 69, 34 000, Kragujevac, Serbia.
Faculty of Engineering, University of Kragujevac, Kragujevac, Serbia.
Int Urol Nephrol. 2025 May 25. doi: 10.1007/s11255-025-04577-0.
Adverse pathological features in prostate cancer (PCa) are characteristics found in biopsy tissue that indicate a more aggressive or advanced disease. This study aims to develop a biopsy-based model for assessing the risk of adverse PCa and to evaluate its performance against the European Randomized Study of Screening for Prostate Cancer (ERSPC) Risk Calculator (RC) 3/4 and PSA models.
Between January 2017 and December 2022, patients with prostate-specific antigen (PSA) levels of ≤ 50 ng/mL underwent prostate biopsies. The patients' age, PSA, digital rectal exam, prostate volume, PSA density (PSAD), previous negative biopsy, number of positive cores, Gleason score, and biopsy outcomes were documented. Patients were classified into categories: no cancer, very low-risk, low-risk, intermediate-risk, and high-risk groups. We investigated the relationship between our model and adverse PCa using a binary Generalized Linear Model (GLM). We evaluated the model's discriminatory ability using the area under the receiver operating characteristic curve (AUC) and compared its predictive performance with the adverse model in terms of discrimination, calibration, and clinical utility.
Out of 824 patients, PCa was diagnosed in 320 (38.8%) men, and 203 (24.6%) had unfavorable PCa. The GLM demonstrated improved performance metrics, with an AUC of 0.766, compared to 0.639 for the RC 3/4 model and 0.655 for the PSA model. The GLM showed a good fit and provided a greater net benefit.
The study identified clinical predictors of adverse PCa during biopsy, demonstrating moderate discrimination and clinical utility. Further large multicenter studies are required for validation.
前列腺癌（PCa）的不良病理特征是在活检组织中发现的特征，表明疾病更具侵袭性或处于更晚期阶段。本研究旨在开发一种基于活检的模型，用于评估不良PCa的风险，并根据欧洲前列腺癌筛查随机研究（ERSPC）风险计算器（RC）3/4和PSA模型评估其性能。
2017年1月至2022年12月期间，前列腺特异性抗原（PSA）水平≤50 ng/mL的患者接受了前列腺活检。记录患者的年龄、PSA、直肠指检、前列腺体积、PSA密度（PSAD）、既往活检阴性、阳性核心数量、Gleason评分和活检结果。患者被分为以下几类：无癌、极低风险、低风险、中风险和高风险组。我们使用二元广义线性模型（GLM）研究了我们的模型与不良PCa之间的关系。我们使用受试者操作特征曲线（AUC）下的面积评估了模型的鉴别能力，并在鉴别、校准和临床实用性方面将其预测性能与不良模型进行了比较。
在824例患者中，320例（38.8%）男性被诊断为PCa，203例（24.6%）患有不良PCa。与RC 3/4模型的AUC为0.639和PSA模型的AUC为0.655相比，GLM显示出更好的性能指标，AUC为0.766。GLM显示出良好的拟合度，并提供了更大的净效益。
该研究确定了活检期间不良PCa的临床预测因素，显示出中等的鉴别能力和临床实用性。需要进一步的大型多中心研究进行验证。
