Eminaga Okyaz, Bögemann Martin, Breil Bernhard, Titze Ulf, Wötzel Fabian, Eltze Elke, Bettendorf Olaf, Semjonow Axel
Department of Urology, University Hospital of Cologne, Cologne, Germany.
Prostate Center, Department of Urology, University Hospital Muenster, Albert-Schweitzer-Campus 1, Muenster, Germany.
Urol Oncol. 2014 Nov;32(8):1317-26. doi: 10.1016/j.urolonc.2014.04.018. Epub 2014 Jun 2.
The prediction value of prostate-specific antigen (PSA) isoform [-2]proPSA (p2PSA) for detecting advanced prostate cancer (PCa) remains unclear. Our objective was to evaluate the additional clinical utility of p2PSA compared with total PSA (tPSA), free PSA (fPSA), and preoperative Gleason score (Gls) in predicting locally advanced PCa (pT3/T4) with high-accuracy discrimination. The aim was to develop a novel classification based on p2PSA and preoperative Gls for predicting advanced PCa.
In 208 consecutive men diagnosed with clinically localized PCa who underwent radical prostatectomy, we determined the predictive and discriminatory accuracy of serum tPSA, fPSA, percentage of fPSA to tPSA, p2PSA, p2PSA density, percentage of p2PSA to fPSA, and the Prostate Health Index. The cutoff level of p2PSA with best accuracy was estimated. The novel classification was developed by analyzing the interaction between p2PSA and Gls in predicting pathologic outcomes using a chi-square automatic interaction detection analysis. Decision curve analysis was applied to test the clinical consequences of using the novel classification.
On univariate analyses, p2PSA, p2PSA density, percentage of p2PSA to fPSA, and Prostate Health Index were accurate but were not independent predictors by multivariate analysis. The p2PSA cutoff level of 22.5 pg/ml showed the best accuracy level for predicting and discriminating advanced diseases (area under the curve [AUC] = 0.725, sensitivity = 51.4%, specificity = 81.8%). By chi-square automatic interaction detection, univariate and multivariate analysis, a p2PSA level > 22.5 pg/ml was significantly associated with an increased frequency and risk of advanced disease. In patients with a p2PSA level ≤ 22.5 pg/ml, 91.8% of Gleason sum 6 PCa was organ confined. The combination of p2PSA and Gls enhanced slightly but significantly the predictive and discriminatory accuracy for advanced disease (0.6%-3.6%).
The p2PSA cutoff level of 22.5 pg/ml can accurately discriminate between organ-confined and advanced PCa. The additional use of p2PSA enhanced slightly the predictive accuracy for advanced PCa (pT3/pT4) and has limited additional predictive value in identifying aggressive PCa (Gls > 7a).
前列腺特异性抗原(PSA)异构体[-2]前体PSA(p2PSA)在检测晚期前列腺癌(PCa)方面的预测价值尚不清楚。我们的目的是评估与总PSA(tPSA)、游离PSA(fPSA)和术前Gleason评分(Gls)相比,p2PSA在高精度鉴别预测局部晚期PCa(pT3/T4)方面的额外临床效用。目的是开发一种基于p2PSA和术前Gls的新型分类方法来预测晚期PCa。
在208例连续诊断为临床局限性PCa并接受根治性前列腺切除术的男性患者中,我们测定了血清tPSA、fPSA、fPSA占tPSA的百分比、p2PSA、p2PSA密度、p2PSA占fPSA的百分比以及前列腺健康指数的预测和鉴别准确性。估计了具有最佳准确性的p2PSA临界值水平。通过使用卡方自动交互检测分析来分析p2PSA和Gls在预测病理结果中的相互作用,从而开发出新型分类方法。应用决策曲线分析来测试使用该新型分类方法的临床后果。
单因素分析中,p2PSA、p2PSA密度、p2PSA占fPSA的百分比以及前列腺健康指数具有准确性,但多因素分析中并非独立预测因子。p2PSA临界值水平为22.5 pg/ml时,在预测和鉴别晚期疾病方面显示出最佳准确性水平(曲线下面积[AUC]=0.725,敏感性=51.4%,特异性=81.8%)。通过卡方自动交互检测、单因素和多因素分析,p2PSA水平>22.5 pg/ml与晚期疾病的频率和风险增加显著相关。在p2PSA水平≤22.5 pg/ml的患者中,91.8%的Gleason总分6分的PCa为器官局限性。p2PSA和Gls的联合使用对晚期疾病的预测和鉴别准确性略有但显著提高(0.6%-3.6%)。
p2PSA临界值水平为22.5 pg/ml可准确区分器官局限性和晚期PCa。p2PSA的额外使用对晚期PCa(pT3/pT4)的预测准确性略有提高,在识别侵袭性PCa(Gls>7a)方面的额外预测价值有限。