Lai Win Shun, Gordetsky Jennifer B, Thomas John V, Nix Jeffrey W, Rais-Bahrami Soroush
Department of Urology, University of Alabama at Birmingham, Birmingham, Alabama.
Department of Pathology, University of Alabama at Birmingham, Birmingham, Alabama.
Cancer. 2017 Jun 1;123(11):1941-1948. doi: 10.1002/cncr.30548. Epub 2017 Jan 31.
The objective of this study was to create a nomogram model integrating clinical and multiparametric magnetic resonance imaging (MP-MRI)-based variables to predict prostate cancer upgrading in a population of active surveillance (AS) patients.
Prostate cancer patients on AS who underwent MP-MRI with magnetic resonance imaging (MRI)/ultrasound (US) fusion-guided biopsy were identified. Clinical and imaging variables, including the prostate-specific antigen density (PSAD), number of lesions, total lesion volume, total lesion density, Prostate Imaging Reporting and Data System magnetic resonance imaging suspicion score (MRI-SS), and duration between prereferral systematic and MRI/US fusion-guided biopsy sessions, were assessed. Logistic regression modeling was used to assess upgrading on MRI/US fusion-guided biopsy. A predictive model for upgrading was calculated with the significant factors identified.
Seventy-six patients were analyzed with a mean age of 62.5 years and a median prostate-specific antigen (PSA) level of 5.1 ng/mL. The average duration between prereferral and MRI/US biopsies was 21 months. Twenty patients (26.32%) were upgraded. The PSAD, duration between prereferral and MRI/US biopsies, MRI-SS, and MRI total lesion density were significantly associated with upgrading. A logistic regression model using these factors to predict upgrading on confirmatory MRI/US fusion biopsy had an area under the curve (AUC) of 0.84, whereas the AUC was 0.69 with PSA alone. On the basis of this model, a nomogram was generated, and using a probability cutoff of 22% as an indication of upgrading, it produced sensitivity, specificity, positive predictive, and negative predictive values of 80%, 81.25%, 57.1%, and 92.86%, respectively.
The integration of MRI findings with clinical parameters can add value to a model predicting upgrading from a Gleason score of 3 + 3 = 6 in men on AS. This can potentially be used as a noninvasive approach to confirm AS patients with low-risk disease for whom biopsy may be deferred. Cancer 2017;123:1941-1948. © 2017 American Cancer Society.
本研究的目的是创建一个整合临床和基于多参数磁共振成像(MP-MRI)变量的列线图模型,以预测主动监测(AS)患者群体中前列腺癌的升级情况。
确定接受MP-MRI检查并经磁共振成像(MRI)/超声(US)融合引导活检的AS前列腺癌患者。评估临床和影像变量,包括前列腺特异性抗原密度(PSAD)、病灶数量、总病灶体积、总病灶密度、前列腺影像报告和数据系统磁共振成像可疑评分(MRI-SS)以及转诊前系统活检与MRI/US融合引导活检之间的时间间隔。使用逻辑回归模型评估MRI/US融合引导活检时的升级情况。根据确定的显著因素计算升级的预测模型。
分析了76例患者,平均年龄62.5岁,前列腺特异性抗原(PSA)水平中位数为5.1 ng/mL。转诊前与MRI/US活检之间的平均时间间隔为21个月。20例患者(26.32%)出现升级。PSAD、转诊前与MRI/US活检之间的时间间隔、MRI-SS以及MRI总病灶密度与升级显著相关。使用这些因素预测确诊MRI/US融合活检时升级的逻辑回归模型的曲线下面积(AUC)为0.84,而仅使用PSA时AUC为0.69。基于该模型生成了列线图,以22%的概率截断值作为升级的指标,其敏感性、特异性、阳性预测值和阴性预测值分别为80%、81.25%、57.1%和92.86%。
将MRI结果与临床参数相结合可为预测AS男性患者从Gleason评分3 + 3 = 6升级的模型增加价值。这有可能用作一种非侵入性方法,以确认可能推迟活检的低风险疾病AS患者。《癌症》2017年;123:1941 - 1948。© 2017美国癌症协会。