PSA

Synopsis

 Glycoprotein;
 kallikrein related serine protease produced by secretory epithelium, drains into ductal system;
 cleaves and liquefies seminal coagulum formed after ejaculation.
 PSA > 4 seen in 80% with histologically documented cancer but also in 25-30% with nodular hyperplasia, prostatitis, infarcts, prostatic massage, cystoscopy; elevated in 2 of 18 post-race marathon runners (Archives 2003;127:345)
 Annual testing recommended for men 50+, men 40+ at increased risk
PSA density (PSA per volume of prostate gland), velocity (changes in PSA with time), %free (unbound to alpha-1-antichymotrypsin), serial measurements important for follow up (AJCP 1994;102 (4 Supp 1): S31)

Prostate specific antigen (PSA)

PSA is produced by the epithelial cells lining the prostatic ducts and acini and is secreted directly into the prostatic ductal system. The PSA gene is located on chromosome 19.

Its androgen-regulated transcription results in the biosynthesis of a 261 amino acid PSA precursor. This precursor, is believed to be activated by the proteolytic liberation of a small amino-terminal fragment.

Conversion from inactive proPSA to active PSA requires action of exogenous prostatic proteases, e.g. hK2, prostin (hK15), prostase (hK4) or trypsin.

Different molecular forms of PSA exist in serum. These result from complex formation between free PSA and two major extracellular protease inhibitors that are synthesized in the liver.

As PSA is a serine protease, its normal mode of existence in the serum is in a complex with α-1-anti-chymotrypsin (ACT), a 67 kDa single chain glycoprotein, and α-2-macroglobulin (AMG), a 720 kDa glycoprotein.

Only a small percentage of the PSA found in the serum is free. Because this free form does not bind to ACT or AMG, it is thought to be either the enzymatically inactive precursor (i.e., zymogen) for PSA or an inactive nicked or damaged form of the native molecule.

Subfractions of free PSA include: mature single-chain, and multichain, nicked free PSA forms.

Prognostic value of serum PSA

PSA is the key factor in the screening for and detection of prostate cancer, its serum level at the time of diagnosis is considered a prognostic marker that stratifies patients into differing prognostic
categories.

Recent reports, however indicate that the prognostic value is driven by patients with high PSA levels, which is significantly associated with increasing tumour volume and a poorer prognosis.

In recent years, however, most newly diagnosed patients have only modestly elevated PSA (between 2 and 9 ng/ml), a range in which BPH and other benign conditions could be the cause of the PSA elevation.

For patients within this category, it was reported that PSA has no meaningful relationship to cancer volume and grade in the radical prostatectomy specimen, and a limited relationship with PSA cure rates.

Following treatment, serum PSA is the major mean of monitoring
patients for tumour recurrence.

Serum total PSA and age specific reference ranges

Serum PSA is determined with immunoassay techniques. No PSA epitopes that interact with anti-PSA antibodies are exposed on the PSA-AMG complex.

This is thought to result from the 25- fold larger AMG molecule "engulfing" PSA and hindering recognition of PSA epitopes.

Therefore, conventional assays do not measure PSA-AMG. In contrast, only one major PSA epitope is completely shielded by complex formation with ACT; PSA-ACT can therefore be readily measured in serum.

Monoclonal antibodies have been designed to detect the free form of PSA (29kDa), the complex of PSA and ACT (90 kDa) and the total PSA.

It has been found that total PSA correlates well with advancing age. Based on the 95th percentile values in a regression model, white men under age 50 have PSA values < 2.5 ng/ml, under age 60 have PSA values < 3.5 ng/ml, under age 70 have PSA values < 4.5 ng/ml, and under age 80 PSA levels were < 6.5 ng/ml. It has been suggested that these age-related values be used as the upper limit of normal in PSA-related diagnostic strategies.

PSA is elevated beyond the arbitrary cutoff point of 4.0 ng/ml in the majority of patients with prostate cancer.

It may also be greater than 4.0 ng/ml in some benign conditions, including benign prostatic hyperplasia (BPH).

Prostate cancer may also be present in men with serum PSA values lower than the above quoted cutoff points. This may be specifically true for men considered at higher risk (i.e., family history; men with faster doubling time; and in the United States African American men).

Therefore, serum PSA lacks high sensitivity and specificity for prostate cancer. This problem has been partially overcome by calculating several PSA-related indices and/or evaluating other serum markers.

PSA tests are also useful to detect recurrence and response of cancer following therapy. The exact value used to define recurrence varies depending on the treatment modality.

Free PSA

The free form of PSA occurs to a greater proportion in men without
cancer and, by contrast, the α-1-chymotrypsin complex PSA comprises a greater proportion of the total PSA in men with malignancy.

The median values of total PSA and of the free-to-total PSA ratio are 7.8 ng/ml and 10.5% in prostate cancer patients, 4.3 ng/ml and 20.8% in patients with BPH, and 1.4 ng/ml and 23.6% in a control group of men without BPH.

There is a significant difference in free-to-total PSA ratio between
prostate cancer and BPH patients with prostate volumes smaller than 40 cm3, but not between patients in these two groups with prostate volumes exceeding 40 cm3.

Complex PSA. Problems associated with the free-to-total PSA ratio, particularly assay variability, and the increased magnitude of error when the quotient is derived, are obviated by assays for complex PSA.

Complex PSA value may offer better specificity than total and free-tototal PSA ratio.

PSA density

This is the ratio of the serum PSA concentration to the volume of the gland, which can be measured by transrectal ultrasound (total PSA/prostatic volume = PSA density, PSAD).

The PSAD values are divided into three categories: normal (values equal or lower than 0.050 ng/ml/cm3), intermediate (from 0.051 to 0.099 ng/ml/cm3) and pathological (equal to or greater than 0.1 ng/ml/cm3).

The production of PSA per volume of prostatic tissue is related to the presence of BPH and prostate cancer and to the proportion of epithelial cells and the histological grade of the carcinoma.

PSA density of the transition zone

Nodular hyperplasia is the main determinant of serum PSA levels in patients with BPH. Therefore, it seems logical that nodular hyperplasia volume rather than total volume should be used when trying to interpret elevated levels of serum PSA.

PSA density of the transition zone (PSA TZD) is more accurate in predicting prostate cancer than PSA density for PSA levels of less than 10 ng/ml.

Prostate-specific antigen epithelial density. The serum PSA level is most strongly correlated with the volume of epithelium in the transition zone. The prostate-specific antigen epithelial density (PSAED, equal to serum PSA divided by prostate epithelial volume as determined morphometrically in biopsies) should be superior to PSAD.

However, the amount of PSA produced by individual epithelial cells is variable and serum levels of PSA may be related to additional factors such as hormonal milieu, vascularity, presence of inflammation, and other unrecognized phenomena.

PSA velocity and PSA doubling time

PSA velocity (or PSA slope) refers to the rate of change in total PSA levels over time. It has been demonstrated that the rate of increase over time is greater in men who have carcinoma as compared to those who do not.

This is linked to the fact that the doubling time of prostate cancer is estimated to be 100 times faster than BPH. Given the shortterm variability of serum PSA values, serum PSA velocity should be calculated over an 18-month period with at least three measurements.

PSA doubling time (PSA DT) is closely related to PSA velocity 1470. Patients with BPH have PSA doubling times of 12 ± 5 and 17 ± 5 years at years 60 and 85, respectively. In patients with prostate cancer, PSA change has both a linear and exponential phase. During the exponential phase, the doubling time for patients with local/regional and advanced/metastatic disease ranges from 1.5-6.6 years (median, 3 years) and 0.9-8.5 years (median, 2 years), respectively.

Pathology

PSA is a glycoprotein produced almost exclusively in the epithelium of the prostate gland. In the circulation PSA may be complexed to serum proteins (complexed PSA, or cPSA) or may be free (fPSA). The cPSA and fPSA together comprise total PSA (tPSA). The tPSA is normally less than 4 ng/mL (normal ranges vary depending upon which assay is used).

A mildly increased tPSA in a patient with a very large prostate can be due to nodular hyperplasia, or to prostatitis, rather than carcinoma. The fPSA correlates more closely with benign prostatic conditions than the tPSA.

The cPSA has a greater sensitivity for prostatic adenocarcinomas at the low ranges of elevation. A rising tPSA is suspicious for prostatic carcinoma, even if the tPSA is in the normal range.

Transrectal needle biopsy, often guided by ultrasound, is useful to confirm the diagnosis, although incidental carcinomas can be found in transurethral resections for nodular hyperplasia. (Jung et al, 2006)

Men who have findings suspicious for carcinoma on digital rectal examination and a tPSA of < 4 ng/mL have a probability of cancer of at least 10%, while those with tPSA levels from 4 to 10 ng/mL have a 25% probability. Men with tPSA’s above 10 ng/mL have a >50% likelihood of having a prostate cancer. (Demura et al, 1996)

Reverse transcriptase-polymerase chain reaction

RT-PCR is an extremely sensitive assay, capable of detecting one prostate cell diluted in 108 non-prostate cells. This high degree of sensitivity mandates that extreme precaution be taken to avoid both
cross-sample and environmental contamination.

Because of the high sensitivity of RT-PCR, there is the possibility that extremely low-level basal transcriptions of prostate-specific genes from non-prostate cells will result in a positive RT-PCR signal.

More recently, basal PSA mRNA levels were detected in a quantitative RT-PCR in individuals without prostate cancer, thus suggesting the need to quantitate the RTPCR assay in order to control for basal transcription. These problems with RT-PCR have limited its clinical utility.