The Human Genome Project, an ambitious effort comprising the sequencing of 23,000 human genes, ultimately opened the door to an even more daunting problem: that of systematically identifying and characterizing the proteins encoded by the genome. Due to post-translational modifications and combinatorial splicing, the number of proteins encoded by the genome is likely more than an order of magnitude greater than the number of human genes. Furthermore, proteomics—the large-scale study of protein structure and function—is made more complex by the dynamic expression of proteins over time and by their interactions in situ. That dynamic nature, however, gives rise to a specific and clinically relevant opportunity: Disease states may well be reflected in the expression of specific proteins as cell nuclei undergo cancer-induced changes.

Such an opportunity, as it applies to the early diagnosis of prostate cancer, exists in the detection of a nuclear matrix protein—early prostate cancer antigen (EPCA) —that is believed to be expressed in the early stages of carcinogenesis.¹ This diagnostic approach differs significantly from prostate specific antigen (PSA) analysis in that the latter relies on a marker specific to the prostate gland, not necessarily specific to neoplastic prostate tissue.

Advances and Limitations of PSA Screening

This year is the 25th anniversary of the discovery of PSA. Controversy, however, still remains regarding the utility of the molecule. Much of this controversy traces back to the initial characterization of the marker. At the time that PSA was discovered, it was clear that it was not prostate cancer specific. The nomenclature for the antigen was appropriate in that it was described as prostate specific antigen and not, prostate cancer specific antigen. PSA is found in the normal prostate at equal or higher levels than it is found in prostate cancer. It is known that diseases of the prostate, like benign prostatic hyperplasia (BPH), prostatitis, as well as prostate cancer, result not necessary in higher prostatic levels of this marker, but in more of the marker being spilled into the serum.

In prostate cancer, one of the reasons that the marker may so easily get into the serum relates to changes in the prostate cancer cells themselves. In a normal prostate, PSA is secreted into the acini or central portion of the prostatic glands where it then empties into the ejaculate. In prostate cancer, cellular organization is altered, and the protein does not end up going where it should. Instead, it often leaks into the surrounding blood vessels causing the protein to be found at higher levels within the blood of these patients. Furthermore, it is known that in prostate cancer the blood vessels are more permeable, making it easier for PSA to get into the bloodstream.

Despite these limitations, PSA screening has changed the course of the disease. When PSA was first discovered, more than half of the men presenting with prostate cancer did so with metastatic disease. Today, this rarely happens. In fact, it appears that prostate cancer is being detected at earlier and earlier stages.

The limitations of PSA, though, still exist and may be more severe than ever. A recent study has shown that approximately 17% of men with “normal” PSA levels, even those with less than 1.0 ng/ml, have cancer cells within their prostate.² Some of these prostate cancers may be very aggressive in that they may have become, “dedifferentiated”—or less like the normal prostate—and, therefore, have turned off expression of PSA. On the other end, we know that a large majority—as high as 78%—of men with what are considered to have elevated levels of PSA, particularly in the range of 4 ng/ml to 9.9 ng/ml do not have prostate cancer.³

A number of tools have been applied to make PSA a better marker. First, PSA has been analyzed on a temporal basis in order to follow the biology of an individual’s prostate. By looking at factors like PSA doubling time, as well as PSA velocity, we can begin to more accurately differentiate those with prostate cancer from those with other benign diseases of the prostate. In addition, this information may be predictive in identifying a group of men who have more aggressive disease.

Research aimed at improving PSA as a biomarker is ongoing. A number of isoforms of PSA may be slightly additive when evaluated with free PSA. Several recent studies indicate that there is some value in looking at the ratio of percent free PSA, as well as complex PSA.

However, PSA has, as yet, not met the challenge of separating the prostate cancers with the potential to kill from those without such potential. This is truly the holy grail of prostate cancer markers.

Elevated PSA - 'Negative Biopsy' Dilemma

As a result of the PSA era, a number of important patient populations have developed. The first of these consists of a large number of men with elevated PSA levels who have had at least one set of biopsies of the prostate—the pathology of which has come back “negative.” There are more than 25 million such cases in the United States today. These men, in general, live in uncertainty about whether they actually have cancer within their prostate that was missed on the biopsy or whether, in fact, they do not have the disease. In addition, every year more men undergo biopsies for prostate cancer. In 2005, it is estimated that there will be 1.8 million men biopsied for prostate cancer, and, of those, 232,000 will be diagnosed with the disease.⁴ Therefore, a large majority of the men undergoing biopsy for the disease do not have prostate cancer. There is clearly a need to better focus screening efforts and to biopsy only those who have a high risk for having the disease.

Focused Proteomics

To address these issues, researchers at the University of Pittsburgh and Johns Hopkins University have sought to develop novel biomarkers for the detection of prostate cancer. To do so, they have used a hypothesis-driven approach termed “focused proteomics.” With such a strategy, comparisons are made between cancer cells and normal cells and differing markers are identified. The markers serve as the molecular correlates of the pathology determining whether there is prostate cancer within the prostate.

Because it is known that alterations in the nucleus, including changes in nuclear shape, are a hallmark of the cancer process, researchers have focused on a set of nuclear-structure proteins. By deciphering the protein components of the nucleus, researchers anticipate being able to delve into the molecular signature of a cancer cell.

Several specific markers of prostate cancer have been identified using this approach. The first of these markers was termed Early Prostate Cancer Antigen or EPCA. Tools designed to detect this marker revealed that it has some very unusual properties. Not only does the prostate cancer itself stain positive for EPCA, but so do the normal, surrounding areas of the prostate. This finding indicates that EPCA appears to detect what is called a “field effect” in the prostate.1 Several investigators have reported that the entire prostate of men with prostate cancers is not normal. Changes happen throughout the gland that may make it unstable and resulting in cancer in many distinct areas within the glands.

The discovery that EPCA was found throughout the prostate of men with prostate cancer led researchers to the next question of interest: Could they tell if a patient had prostate cancer, even if a biopsy were sampled at the normal areas of the prostate?

A study by Dr. Rajiv Dihr at the University of Pittsburgh revealed that, based upon evaluation of “normal tissue,” EPCA does indeed separate those with prostate cancer from those without the disease.⁵ The analysis went back and looked at over 300 negative biopsies from men who were later diagnosed with prostate cancer. In this population, men were identified who had negative biopsies up to five years earlier. These tissue biopsies were indistinguishable by normal pathologic analysis, but the EPCA stain revealed the presence of prostate cancer in those with the disease. The sensitivity of the test was 84%. Its specificity was 85%.

These initial findings have been validated in a separate study in which approximately 94% of individuals with prostate cancer were identified by looking for EPCA staining in the normal areas of the prostate.1 Also of note was the fact that EPCA was positive in proliferative inflammatory atropy (PIA) and prostatic intraepithelial neoplasia (PIN) lesions among 86% of the prostate cancer patients.

Conclusion

Based on a hypothesis related to nuclear biomarkers, a prostate cancer specific marker is now in clinical use. This marker can be used to stain biopsies for men and reveal whether prostate cancer may exist within the gland.

This data should help guide both the patients and their urologists in deciding whether repeat biopsies should be done and when. The current consensus of researchers is that if EPCA staining is positive in a biopsy that otherwise appears normal, a repeat biopsy should be performed. It is almost a certainty that there is prostate cancer within the gland. Men whose biopsies are negative for EPCA should feel reasonably safe that they do not have the disease; however, they should continue typical monitoring of their progress. It is hoped that further development of an EPCA biomarker for prostate cancer will aid the large number of men who have elevated PSA levels but \negative biopsies.

A question for further exploration will be whether a relationship exists between the expression of EPCA in the tumor and the aggressiveness of the malignancy.

References

² Thompson IM, Pauler DK, Goodman PJ, et al. Prevalence of prostate cancer among men with a prostate-specific antigen level < or =4.0 ng per milliliter. N Engl J Med. 2004;350:2239-2246.

³ Catalona WJ, Smith DS, Ratliff TL, et al. Measurement of prostate-specific antigen in serum as a screening test for prostate cancer. N Engl J Med. 1991;324:1156-1161.

⁴ American  Cancer Society. Cancer Facts and Figures 2005. Available at http://www.cancer.org/downloads/STT/CAFF2005f4PWSecured.pdf. Accessed October 21, 2006.

⁵ Dhir R, Vietmeier B, Arlotti J, et al. Early identification of individuals with prostate cancer in negative biopsies. Journal of Urology.2004;171:1419-1423.