Toubert M E, Guillet J, Chiron M, Meria P, Role C, Schlageter M H, Francois H, Borschneck C, Nivelon F, Desgrandchamps F, Rastel D, Cussenot O, Teillac P, Le Duc A, Najean Y
Nuclear Medicine Department, Saint-Louis Hospital, Paris, France.
Eur J Cancer. 1996 Nov;32A(12):2088-93. doi: 10.1016/s0959-8049(96)00245-6.
Prostate-specific antigen (PSA) is a protease able to bind to serum antiproteases as alpha 1 antichymotrypsin (ACT). Free PSA (FPSA) corresponds to the fraction of total PSA (TPSA) which is unbound to ACT. Specific detection of the FPSA seems to be a valuable tool in the distinction between prostatic cancer (PCa) and benign prostatic hyperplasia (BPH). Our aim was to evaluate retrospectively the FPSA/TPSA ratio in comparison to TPSA or FPSA determination, using two new immunoradiometric assays (PSA-RIACT and FPSA-RIACT, CIS bio international, Gif Sur Yvette, France) in the early diagnosis of PCa. 256 men, with TPSA levels between 0.7 and 44.7 ng/ml (median age = 69 years), including 164 sera obtained from patients with BPH and 92 sera from patients with untreated PCa were assayed. All diagnoses were histologically confirmed and patients tested before any adjuvant treatment. The evaluation of the median FPSA/TPSA ratio in the two groups showed significantly different values (BPH group: 24.2%, PCa group: 12.1%, P < 0.0001). By R.O.C. (Receiver-Operating-Characteristics) analysis, we show that the FPSA/TPSA ratio is the method of choice for discriminating BPH and PCa, since the area under curve is the greatest for the FPSA/TPSA ratio curve, as compared to the TPSA or FPSA curves (P < 0.0001). The best accuracy (number of true positive + true negative/total = 82.4%) was obtained with a FPSA/TPSA ratio < or = 15% with high odds ratio (20.5; confidence interval (CI): 11.2; 37.7). Of interest, similar results were also confirmed even in the subpopulation with serum TPSA levels between 2.5 and 10 ng/ml (161 patients including 99 BPH and 62 PCa). We thus confirm that combined serum measurement of FPSA and TPSA is of particular interest in the early diagnosis of PCa for patients with non-suspicious digital rectal examination and a TPSA value between 2.5 and 10 ng/ml. In those patients, biopsy should be reserved to the cases with FPSA/TPSA below 15%, which allows significant odds ratio (12.8; CI: 5.2; 31.4). Otherwise, to avoid the risk of missing any PCa, usual follow-up with combined TPSA and FPSA determination would be required with the same criteria of biopsy (i.e. FPSA/TPSA ratio < or = 15% when TPSA value is between 2.5 and 10 ng/ml; or TPSA > 10 ng/ml).
前列腺特异性抗原（PSA）是一种蛋白酶，能够与血清抗蛋白酶如α1抗糜蛋白酶（ACT）结合。游离PSA（FPSA）相当于总PSA（TPSA）中未与ACT结合的部分。FPSA的特异性检测似乎是区分前列腺癌（PCa）和良性前列腺增生（BPH）的一项有价值的工具。我们的目的是使用两种新的免疫放射分析方法（PSA-RIACT和FPSA-RIACT，CIS bio international公司，法国伊夫特河畔吉夫），回顾性评估FPSA/TPSA比值与TPSA或FPSA测定相比，在PCa早期诊断中的价值。对256名男性进行了检测，他们的TPSA水平在0.7至44.7 ng/ml之间（中位年龄 = 69岁），其中包括164份来自BPH患者的血清和92份来自未经治疗的PCa患者的血清。所有诊断均经组织学证实，且患者在接受任何辅助治疗前进行检测。两组中FPSA/TPSA比值中位数的评估显示出显著不同的值（BPH组：24.2%，PCa组：12.1%，P < 0.0001）。通过ROC（受试者工作特征）分析，我们发现FPSA/TPSA比值是区分BPH和PCa的首选方法，因为与TPSA或FPSA曲线相比，FPSA/TPSA比值曲线下的面积最大（P < 0.0001）。当FPSA/TPSA比值≤15%时，获得了最佳准确性（真阳性数 + 真阴性数/总数 = 82.4%），且优势比高（20.5；置信区间（CI）：11.2；37.7）。有趣的是，即使在血清TPSA水平在2.5至10 ng/ml之间的亚组（161名患者，包括99名BPH患者和62名PCa患者）中，也证实了类似的结果。因此，我们证实，对于直肠指检无异常且TPSA值在2.5至10 ng/ml之间的患者，联合检测血清FPSA和TPSA在PCa早期诊断中具有特别重要的意义。在这些患者中，活检应仅保留给FPSA/TPSA低于15%的病例，其优势比显著（12.8；CI：5.2；31.4）。否则，为避免漏诊任何PCa的风险，对于相同的活检标准（即当TPSA值在2.5至10 ng/ml之间时FPSA/TPSA比值≤15%；或TPSA > 10 ng/ml），需要进行常规的TPSA和FPSA联合检测随访。
