Evolving Role of Pro-PSA as a New Serum Marker for the Early Detection of Prostate Cancer

The development and subsequent routine use of prostate-specific antigen (PSA) over the past decade has revolutionized the management of prostate cancer. PSA has increased our ability to detect and, in turn, treat early prostate cancer; however, the major drawback of PSA is its relative lack of specificity. This is especially important in the critical diagnostic range of 4–10 ng/mL, where an elevated PSA may reflect either prostate cancer or benign disease, such as benign prostatic hyperplasia (BPH). This lack of specificity has led to unnecessary prostate biopsies, with associated anxiety, cost, and potential morbidity.

It was discovered in 1991 that serum contains two distinct major forms of PSA: one form is covalently bound to endogenous serum protease inhibitors like α₁-antichymotrypsin and is known as complexed PSA; the other form is present in a “free,” nonactive, noncomplexed form and is known as free PSA.^1,2 The measurement of the ratio of free to total PSA has led to a modest but significant improvement in the discrimination of prostate cancer from BPH in men with PSA levels between 4.0 and 10.0 ng/mL. This is attributable to the association of BPH with high levels of free PSA, compared with prostate cancer.³ The free PSA form is a heterogeneous group consisting of at least three different subforms of inactive PSA. One form has been identified as the pro-enzyme or precursor form of PSA (pPSA), which contains a seven-amino-acid pro-leader peptide (APILSR) and is associated with prostate cancer.^4,5 PSA is normally found in seminal fluid in a mature, active form with 237 amino acids, lacking the pro-leader peptide. The pro-leader peptide is removed extracellularly by human kallikrein 2 (hK2) to produce this active, mature form of PSA. Intracellularly, the pPSA form is itself truncated as a result of post-translational proteolytic cleavage of the pro-leader peptide, where either three or five of the seven pro-leader amino acids are removed. All forms of truncated pPSA (ie, [−4] and [−2]) remain enzymatically inactive. It is these truncated forms of pPSA that have recently generated a great deal of clinical interest, as they may have a diagnostic role in the early detection of prostate cancer. The following recent articles from Cancer Research investigate and discuss this possibility.

A Precursor Form of Prostate-Specific Antigen Is More Highly Elevated in Prostate Cancer Compared with Benign Transition Zone Prostate Tissue

Mikolajczyk SD, Millar LS, Wang TJ, et al.

Cancer Res. 2000;60:756–759.

Having previously reported that pPSA is a component of free PSA in the serum of prostate cancer patients, these authors examined matched sets of tissues harvested from patients undergoing radical prostatectomy (n = 18). From each prostate, samples of prostate cancer and adjacent, noncancerous peripheral-zone tissues were selected for analysis. Furthermore, a sample (n = 8) of benign transitional-zone tissue, obtained from transurethral resection of prostates, was also analyzed. PSA was immunoaffinity purified from these prostate tissues, and the authors found that pPSA was differentially elevated in the peripheral zone of cancer tissue and was largely undetected in the transition-zone tissue. N-terminal sequencing revealed that the pPSA was composed primarily of the truncated [−2]pPSA with minor levels of [−4]pPSA. The median value of pPSA was 3% in peripheral zone of cancer tissue and 0% (undetectable) in the transitional zone (P < .0026). pPSA was not detected in 13 of the 18 transitional-zone tissue specimens (72%). Of the 18 matched cancer specimens, 16 (89%) contained measurable pPSA. The authors conclude that pPSA is more highly correlated with prostate cancer than with BPH. In addition, pPSA in serum may represent a more cancer-specific form of PSA that could help to distinguish prostate cancer from BPH, especially in patients with only mildly elevated PSA.

Identification of Precursor Forms of Free Prostate-Specific Antigen in Serum of Prostate Cancer Patients by Immunosorption and Mass Spectrometry

Peter J, Unverzagt C, Krogh TN, et al.

Cancer Res. 2001;61:957–962.

The authors obtained serum from 5 patients with prostate cancer and subsequently isolated free PSA by immunopurification procedures using streptavidin-coated magnetic beads. They then identified pPSA forms using matrix-assisted laser desorption ionization time-of-flight mass spectrometry, after producing peptides by endoproteinase from Lysobacter enzymogenes digestion of the SDS poly-acrylamide gel electrophoresis-separated free PSA bands. They found that among the five serum samples investigated, all contained the [−7], [−5], and [−4] pro-PSA forms, whereas the [−1] and [−2] forms were only present in three of them. The authors, therefore, were able to demonstrate that sera obtained from prostate cancer patients have the pro-PSA forms and occur in various combinations. The results, however, differed from those of Mikolajczyk and colleagues (reviewed above), who identified the [−4] and [−2] forms of pPSA in tissue extracts but did not find the longer precursor sequences (ie, [−7] or [−5]) forms of pPSA. The reason for this difference may reflect mere sample variability in a small population, or it may reflect the fact that these authors used serum with much higher serum PSA values (one patient’s PSA was 1890 ng/mL, and the other four samples had PSA values > 6000 ng/mL). In contrast, the serum by Mikolajczyk and colleagues had much lower serum PSA values.

A Truncated Precursor Form of Prostate—Specific Antigen is a More Specific Serum Marker of Prostate Cancer

Mikolajczyk SD, Marker KM, Millar LS, et al.

Cancer Res. 2001;61:6958–6963.

The authors have previously identified that the [−2]pPSA truncated form of pPSA is selectively present in prostate cancer tissues and have developed monoclonal antibodies to detect [−2]pPSA and other isoforms of pPSA for Western blot analysis. PSA was immunoaffinity purified from 100–200 mL of serum from five men with biopsy-proven prostate cancer (PSA range: 6–24 ng/mL; mean: 13.4 ng/mL) and three biopsy-negative patients (PSA range: 7–12 ng/mL; mean: 9.7 ng/mL). The truncated [–2]pPSA was found to range from 25% to 95% of the free PSA in the five cancer samples; however, in the three biopsy-negative samples, this value was only 6%–19%. Immunohistochemical studies showed positive staining for [–2]pPSA in prostate cancer tissue epithelium and that [–2]pPSA was enriched in cancer cell secretions, thereby further strengthening the view that [–2]pPSA is naturally present in prostate tissues and is not the artifactual result of tissue extraction methodologies. The authors also performed in vitro activation studies, which revealed that hK2 and trypsin readily activated full-length pPSA by release of the pro-leader peptide but were unable to activate [–2]pPSA to mature PSA. Hence, once formed, [–2]pPSA appears to be a stable but inactive isoform of PSA. From this work, the authors have concluded that truncated [–2]pPSA may represent an important new diagnostic marker for the early detection of prostate cancer. The authors also reported preliminary results of an unpublished study that serum [–2]pPSA levels were threefold higher in 20 biopsy-positive patients compared to 20 biopsy-negative patients with total serum PSA levels between 2 and 22 ng/mL.

The investigations reviewed here are promising and may lead to the development of pPSA as a new marker for the early detection of prostate cancer. Clinical studies with much larger patient population are required, however. In addition, as free PSA constitutes only a small percentage of total PSA, at present, large volumes of serum are required to obtain pPSA information. This may negatively impact pPSA as a potential new marker. Further knowledge of the role of [–2]pPSA will enable us to perform comprehensive trials of pPSA in prostate cancer early detection.