目的:本研究旨在开发一种新的列线图,以预测接受多参数前列腺MRI辅助病变活检的患者的临床上有意义的前列腺癌,解决PI-RADS3病变患者活检的挑战,以及PI-RADS4或5病变阴性患者的随访策略。
方法:使用土耳其泌尿肿瘤协会数据库(UROCaD)进行回顾性病例对照研究。最终的数据集包括2428个病变活检数据。单变量分析,逻辑回归,并进行了验证,在训练和验证数据集中有1942和486个病变活检数据,分别。
结果:年龄,初始总PSA值,PSA密度,前列腺体积,病变长度,DRE发现,和PI-RADS评分在良性或非显着性前列腺癌组和临床意义前列腺癌组之间存在显着差异。开发的列线图包含PSA密度,年龄,PI-RADS评分,病变长度,和DRE调查结果。6倍交叉验证的平均曲线下面积为0.836,而训练和验证数据集的曲线下面积值分别为0.827和0.861。列线图显示在24.9分时的敏感性为75.6%,特异性为74.8%,阳性和阴性预测值分别为42.2%和92.6%。分别。
结论:TUA列线图,基于PSA密度,年龄,PI-RADS评分,病变长度,和DRE的发现,提供了一种可靠和准确的预测工具,用于检测接受多参数前列腺MRI辅助病变(融合)活检的患者的临床显着前列腺癌,可能改善患者管理并减少不必要的活检。
OBJECTIVE: This study aimed to develop a novel nomogram to predict clinically significant prostate cancer in patients undergoing multi-parametric prostate MRI-assisted lesion biopsies, addressing the challenges in deciding on biopsy for patients with PI-RADS 3 lesions and follow-up strategies for patients with negative PI-RADS 4 or 5 lesions.
METHODS: A retrospective case-control study was conducted using the Turkish Urooncology Association Databases (UROCaD). The final dataset included 2428 lesion biopsy data. Univariate analysis, logistic regression, and validation were performed, with 1942 and 486 lesion biopsy data in the training and validation datasets, respectively.
RESULTS: Age, initial total PSA value, PSA density, prostate volume, lesion length, DRE findings, and PI-RADS score were significantly different between benign or non-significant cancer and clinically significant prostate cancer groups. The developed nomogram incorporated PSA density, age, PI-RADS score, lesion length, and DRE findings. The mean area under the curve for the 6-fold cross-validation was 0.836, while the area under the curve values for the training and validation datasets were 0.827 and 0.861, respectively. The nomogram demonstrated a sensitivity of 75.6% and a specificity of 74.8% at a cut-off score of 24.9, with positive and negative predictive values of 42.2% and 92.6%, respectively.
CONCLUSIONS: The TUA nomogram, based on PSA density, age, PI-RADS score, lesion length, and DRE findings, provides a reliable and accurate prediction tool for detecting clinically significant prostate cancer in patients undergoing multi-parametric prostate MRI-assisted lesion (fusion) biopsies, potentially improving patient management and reducing unnecessary biopsies.