Khan Masood A, Partin Alan W, Rittenhouse Harry G, Mikolajczyk Stephen D, Sokoll Lori J, Chan Daniel W, Veltri Robert W
James Buchanan Brady Urological Institute, The Johns Hopkins University School of Medicine, 600 North Wolfe Street, Baltimore, MD 21287, USA.
J Urol. 2003 Sep;170(3):723-6. doi: 10.1097/01.ju.0000086940.10392.93.
In contemporary screening populations a major drawback of prostate specific antigen (PSA) is its relative lack of specificity, especially in the range of 4 to 10 ng/ml, where prostate cancer is found 25% of the time. ProPSA is a derivative of free PSA (fPSA) consisting of the truncated forms (eg [-2]proPSA, [-4]proPSA or the full-length [-7]proPSA). There is increasing evidence that proPSA is associated preferentially with prostate cancer. The objective of this study was to determine whether proPSA can influence the detection of early prostate cancer.
Archival serum samples obtained from 93 men who underwent a systematic 12-core prostate biopsy (total PSA range 4.0 to 10.0 ng/ml) were assayed for percent free PSA, total PSA and the 3 forms of proPSA (Hybritech Tandem Assays Beckman Coulter Access, Beckman Coulter, Inc., Brea, California). Free PSA, the cumulative sum of individual proPSA forms ([-2], [-4] and [-7], or sum-proPSA) and derivatives were determined. Of the 93 men assessed 41 (44%) had evidence of prostate cancer (76% Gleason 5/6, 19% Gleason 7 and 5% Gleason 8). Prostate volume was measured at systematic 12-core biopsy for the detection of prostate cancer. Results were analyzed using univariate and multivariate logistic regression (LR) nonparametric statistical methods.
Using univariate LR, fPSA, percent fPSA (%fPSA), percent sum-proPSA and prostate volume significantly (p <0.05) differentiated men with prostate cancer from those with benign disease. However, applying stepwise backward multivariate LR, total PSA, %fPSA and sum-proPSA were retained and generated a receiver operator characteristic curve with an area under the curve of 76.6%. At 90% sensitivity these 3 variables collectively achieved a specificity of 44% for the detection of prostate cancer. Individually, the 3 retained variables had a specificity of 23% (total PSA), 33% (%fPSA) and 13% (sum-proPSA).
Sum-proPSA, total PSA and %fPSA in combination improve the specificity of early prostate cancer detection in men with a total PSA of 4 to 10 ng/ml compared with the results of individual PSA molecular forms measured.
在当代筛查人群中,前列腺特异性抗原(PSA)的一个主要缺点是其相对缺乏特异性,尤其是在4至10 ng/ml范围内,在此范围内25%的病例被发现患有前列腺癌。前列腺特异性抗原(ProPSA)是游离PSA(fPSA)的衍生物,由截短形式组成(例如[-2]ProPSA、[-4]ProPSA或全长[-7]ProPSA)。越来越多的证据表明,ProPSA与前列腺癌优先相关。本研究的目的是确定ProPSA是否会影响早期前列腺癌的检测。
对93名接受系统性12针前列腺穿刺活检的男性(总PSA范围为4.0至10.0 ng/ml)的存档血清样本进行游离PSA百分比、总PSA以及3种ProPSA形式(Hybritech串联检测,贝克曼库尔特Access,贝克曼库尔特公司,加利福尼亚州布雷亚)的检测。测定游离PSA、各个ProPSA形式([-2]、[-4]和[-7],即总ProPSA)的累积总和及其衍生物。在评估的93名男性中,41名(44%)有前列腺癌证据(76% Gleason 5/6,19% Gleason 7,5% Gleason 8)。在系统性12针前列腺穿刺活检时测量前列腺体积以检测前列腺癌。使用单变量和多变量逻辑回归(LR)非参数统计方法分析结果。
使用单变量LR,fPSA、fPSA百分比(%fPSA)、总ProPSA百分比和前列腺体积能显著(p <0.05)区分前列腺癌男性与良性疾病男性。然而,应用逐步向后多变量LR,总PSA、%fPSA和总ProPSA被保留,并生成了曲线下面积为76.6%的受试者操作特征曲线。在90%的敏感度下,这3个变量共同实现了检测前列腺癌44%的特异性。单独来看,保留的3个变量的特异性分别为23%(总PSA)、33%(%fPSA)和13%(总ProPSA)。
与单独测量的PSA分子形式的结果相比,总ProPSA、总PSA和%fPSA联合使用可提高总PSA为4至10 ng/ml男性早期前列腺癌检测的特异性。