Markers for Prostate Cancer: What's New
by William J. Catalona, M.D.




New findings are improving the screening procedures for prostate cancer. This article is based upon a State-of-the-Art Presentation given by Dr. Catalona at the American Urological Association Annual Meeting this May.




The genetics revolution will change the way prostate cancer is diagnosed and treated.

New technology is beginning to show us the intricate connection between the overexpression of certain genes and the presence of cancer in the prostate.

These new discoveries are in the preliminary stage of research and they show great hope for the diagnosis and treatment of prostate cancer.

Also, more traditional methods for diagnosing prostate cancer are being researched and these new findings are improving the screening procedures.



PSA Still Most Useful Marker

But for the present, the most useful marker for prostate cancer is still PSA. And the research studies in the area of PSA show remarkable improvements in PSA as a diagnostic tool for prostate cancer.

PSA is produced by the epithelial cells lining the glands and ducts of the prostate. It is secreted into the semen in high concentrations but into the blood in low concentrations.

In addition, when the prostate is diseased, as with BPH (benign prostatic hyperplasia, commonly referred to as enlarged prostate) or with cancer, PSA from the prostate seeps into the blood stream, raising PSA levels.

PSA exists in the blood in the form of PSA attached to proteins (also called complexed PSA) and free PSA, not attached and floating free in the blood.

In general, with cancer, more PSA is attached to proteins and therefore its concentration in the blood increases. At the same time, and because more PSA is bonding to proteins, the concentration of free PSA lowers.



Recommendations For Screening

Controversy exists regarding the age patients should begin prostate cancer screening, the appropriate screening intervals, and the PSA threshold to use for biopsy recommendations.

With regard to age, the American Cancer Society and the American Urologic Association recommend the start of screening at age 50 in the general population and at age 40-45 in various high-risk groups such as African- American men and men with a family history of prostate cancer.

With regard to how often screening should be performed, the American Cancer Society and the AUA recommend it take place annually. Others have suggested less intensive screening, such as every two or four years, for young patients and patients with low PSA levels.

A recent analysis suggested, though, that screening at two and four year intervals would result in a large proportion of patients having substantial delays in cancer detection; however, the extent to which these delays might affect outcomes is undetermined.


"...remarkable improvements in PSA as a diagnostic tool"

With regard to PSA threshold for biopsy, traditionally, it was believed that if the serum PSA was between 4 and 10, the chance of detecting prostate cancer on biopsy was 25%. However, we have learned that the proportion of men with detectable cancers in that range is closer to 35 to 40%.

Also, we know that at least 20% of men with readily detectable cancers have PSA levels less than 4, and most of these can be detected earlier by using lower PSA cutoffs.

False positive PSA tests have stimulated a search for methods to reduce unnecessary biopsies without unduly sacrificing cancer detection. These methods have been only partially successful.

The free and complexed PSA readings have been helpful in this regard because we know that with cancer, more PSA bonds to proteins and the percentage of free PSA is lower. In patients whose total PSA is in the diagnostic gray zone of 4-10, taking separate measurements of the PSA and free PSA can help discriminate between cancer and BPH.

For instance, using a cutoff of 25% free PSA for recommending biopsy, 95% of the cancers can be detected while avoiding 20% of unnecessary biopsies.



PSA Screening Methods

While total and free PSA testing are the most useful diagnostic tool in prostate cancer now, new studies are revealing other screening methods which have the potential to improve diagnosis and treatment.

One such method to assist diagnosis is measuring PSA velocity, which helps when the PSA level rises steadily. (PSA velocity measures the increment in PSA values over time telling a doctor how fast the PSA is increasing.) But PSA levels tend to fluctuate and velocity measurements take time to show that the cancer is progressing.

PSA density is helpful as well, but density measurements require ultrasound procedures. PSA density is the PSA value divided by the volume of the prostate, which means than a man with a large gland can expect to have a higher PSA than a man with a small gland.

Also, age-specific reference ranges are useful in men under the age of 50, but they delay the diagnosis of cancer in older men and may compromise their chances for cure.


"Free PSA readings help in reducing unnecessary biopsies."

Another marker for prostate cancer screening is HK2, an enzyme in the PSA family that may have something to do with converting free (inactive) PSA to bound (active) PSA.

The higher the percentage of free PSA in the bloodstream, the less likely of prostate cancer.

Even though HK2 is present in the blood in very low concentrations, those concentrations do increase with the presence of cancer.

Because the percentage of HK2 is higher and the percentage of free PSA is lower when cancer is present, a ratio of the two help in distinguishing between BPH and cancer, especially in men with PSA levels of 2-4.

Preliminary studies suggest that HK2 may be helpful in predicting tumor stage and Gleason patterns.



Pro-PSA Testing

Other PSA isoforms (different molecular forms of PSA) have potential use for telling the difference between BPH and prostate cancer.

Among the isoforms of free PSA are B-PSA, associated with BPH, and pro-PSA, associated with prostate cancer. Currently, BPSA is being evaluated as a potential marker to predict the response to BPH drug therapy.

Pro-PSA levels are higher in cancer patients. Preliminary studies show that in men with total PSA levels between 2.5 and 4, the % of proPSA is a better diagnostic tool than the % of free PSA.

In the PSA range of 4 to 10, the two readings perform comparably. However in men with higher free PSA levels, the % of proPSA is of significantly greater use than the % of free PSA.

In men with greater than 25% free PSA, that would be men who might not normally be recommended to undergo biopsy, the % of proPSA may be able to detect most cancers while avoiding two-thirds of unnecessary biopsies.



Gene Tests

Finally, the emerging technology for detecting gene expression has resulted in the development of gene chips that can register tens of thousands of individual genetic expressions simultaneously on one chip.

In reports from several research groups that have studied gene expression profiling in prostate cancer, hepsin shows up at the top of virtually every list of genes that are over-expressed.

Hepsin is intriguing because it exists on the cell surface and may activate a growth factor that causes prostate cancer cells to multiply. Moreover, mice bred without the hepsin protein are otherwise normal. This finding means that, if an inhibitor of hepsin were developed, it should not have major side effects for the patient.

With the expertise that pharmaceutical chemists currently have in creating protease inhibitors, the discovery of hepsin may not only be useful in identifying patients genetically susceptible to prostate cancer, but also may provide a possible target for prostate cancer therapy and chemoprevention.

This approach is being actively pursued by pharmaceutical companies.


"The genetics revolution will change the way prostate cancer is diagnosed and treated."

Another gene overexpressed in prostate cancer is AMACR (alpha-methylacyl coenzyme A racemase). Preliminary studies suggest that AMACR may have potential for diagnosing prostate cancer in prostate needle biopsy specimens in which the diagnosis of cancer is difficult to make by standard criteria.

Recently, research groups identified a gene on chromosome 1, called RNase L, that is linked to prostate cancer susceptibility in some families.

This gene had been previously cloned and extensively studied, not knowing that it was related to prostate cancer. A genetic variant of this gene that has less enzyme activity than the standard RNase L is statistically associated with prostate cancer.

The implications of these findings are that blood tests for this genetic variant could identify susceptible men and that methods of restoring the enzyme activity might reduce the risk for prostate cancer.

These early examples show the promise of the genetics revolution. In the future, they and many other genes will be studied as potential clinical markers for prostate cancer.