Prostate-Specific Antigen Test: Is It a Valid Screening Tool for Prostate Cancer?

Pamela C. Cartright, MAEd, RT(R)(T)

   ^*Director, Radiation Therapy Program, Diagnostic and Therapeutic Sciences, University of Alabama at Birmingham, Birmingham, Alabama.
   Address correspondence to: Pamela C. Cartright, MAEd, RT(R)(T), Director, Radiation Therapy Program, Diagnostic and Therapeutic Sciences, University of Alabama at Birmingham, RMSB 436, 1705 University Boulevard, Birmingham, AL 35294. E-mail: pamcartr@uab.edu.

ABSTRACT 

Prostate cancer is the second most common form of cancer among men, following only skin cancer. Early detection significantly increases the likelihood of long-term survival. However, the widely used prostate-specific antigen test (PSA), a blood serum test used to detect prostate cancer, is the subject of much debate. Although the incidence of prostate cancer has decreased in recent years, it is unclear how much of this effect can be attributed to PSA screening. Experts disagree on the need for the test, in addition to its effectiveness. Elevated PSA levels may indicate changes in the prostate unrelated to malignancy, producing patient anxiety and prompting physicians to prescribe treatment for conditions that might never be life-threatening. Professional societies, such as the American Cancer Society and the American Urologic Association, have developed different guidelines regarding the role of PSA assessment in cancer screening. Refinements to the PSA test have been made over the past 2 decades, but more research is needed to develop a tumor marker that would better identify aggressive cancers.

Introduction

Although there is consensus that prostate cancer is a major public health problem, there is ongoing controversy about the effectiveness of and the need for prostate-specific antigen (PSA) testing to detect prostate cancer. A review of current literature presents 2 distinct approaches to the available data. Advocates for PSA screening, such as Oottamasathien and Crawford, point to falling mortality rates in the United States and an increase in the use of PSA screenings to argue that the tests are responsible for saving lives.¹ Despite frequent false positives, they advocate combined PSA screenings and digital rectal examination (DRE); reasoning that, by using these tests, more tumors are detected.¹ However, this position is disputed by other experts. Hoffman thinks that tumor detection is not the only important criterion, but that the patient's quality of life, and potentially irreversible side effects of early (and possibly unnecessary) treatment, must be also considered.² Hoffman takes a more conservative approach to his research. He criticizes advocates for using only 1 randomized screening trial, the results of which demonstrated benefit over a period of only 3 years, when the "slow-growing cancer" usually has a "5-year lead time."² He points out that the evidence supporting the benefit of PSA screening is only indirect and suggests waiting for more concrete information from studies. He also expresses concern that advocates are not taking into account other advances in treatment and palliation that may have an effect on death rate data.²

According to the American Cancer Society, the most common form of cancer found in men (excluding skin cancer) is prostate cancer.³ They estimate that approximately 232 000 new cases are diagnosed each year in the United States.³ Although prostate cancer is typically slow growing, there are more aggressive forms of the disease, and the number of deaths from this type of cancer is second only to lung cancer.³ Autopsy studies have underscored the pervasiveness of the disease, revealing that approximately 1 in 3 has histological evidence of prostate cancer by age 50 years.¹ Autopsies of men in their 80s show that more than 70% have metastatic prostate cancer, and greater than 90% demonstrate benign prostatic hyperplasia (BPH).⁴

Survival rates vary according to the stage at which the disease is detected. The American Cancer Society reports that approximately 86% of prostate cancers are detected before metastasizing to other organs in the body, and that in such cases the 5-year survival rate is nearly 100%.³ However, when the disease is not detected until distant metastasis has occurred, only 34% will survive 5 years.³ In many of these cases, late-stage prostate cancer can be treated only palliatively.² This demonstrates the need for early detection and, clearly, accurate screening techniques are essential. Currently employed screening techniques include the DRE, PSA, transrectal ultrasound (TRUS), and biopsy.⁵ Although the usefulness of the DRE, TRUS, and biopsy are largely undisputed, the use of the PSA to detect prostate cancer is in question by leading experts.

How PSA Works

Prostate-specific antigen is a protein produced and secreted primarily by the prostatic epithelium. The exact process by which PSA enters blood circulation remains unknown, but its presence can be detected by blood serum tests developed in 1979 by T. Ming Chu.⁶ When prostate cancer develops, the PSA level usually goes above 4 ng/mL. If the PSA level is on the borderline range (between 4 and 10 ng/mL), the patient is thought to have approximately a 25% chance of having prostate cancer. If the PSA level is above 10 ng/mL, the chance of having prostate cancer increases to more than 67% (Table 1).^7,8 PSA levels are slightly different for African American and Asian males as compared to white males. Although there are studies to support the findings that the normal value is higher for African American men, PSA values in these men are not well defined.⁹ According to a study by Henderson et al, "Published criteria for normal PSA level and density have been derived primarily from white men and may not be directly applicable to other populations. Race-specific data are needed to fully optimize PSA as a tumor marker in racial populations that are at high risk for prostate cancer death."⁹

Prostate-specific antigens can be affected by other factors. PSA levels increase as a result of noncancerous enlargement of the prostate, known as BPH, and prostatitis (inflammation of the prostate). PSA levels naturally increase with age and after ejaculation. In addition, some herbal remedies may alter blood PSA levels.⁵

Because of the high incidence of false positives, urologists have been working to refine the PSA test since the mid 1980s.¹⁰ Some methods of improving accuracy include determining age-specific PSA ranges (because PSA levels normally increase with age), PSA velocity (using sequential screenings to determine the rate at which levels are increasing), percent-free PSA (which indicates how much PSA circulates freely in the bloodstream vs the total PSA level; Table 2), PSA density (used for men with large prostate glands), and the ratio of PSA/IGF-1 (insulin-like growth factor-1; to distinguish patients with BPH from those with cancer).^5,10,11

As a follow-up test, percent-free prostate-specific antigen (free PSA) may be used to see if a prostate biopsy should be done to check for cancer.⁶ Men with a low percentage of free PSA have a higher chance of developing cancer, as shown in Table 2.¹¹

A Controversial Test

Although there is consensus that prostate cancer is a major public health problem, there is ongoing controversy about the effectiveness of and the need for PSA testing to detect prostate cancer. A counter opinion has been provided by Oottamasathien and Crawford that PSA testing is "the most important tumor marker for the detection, staging, and monitoring of men with prostate cancer."¹ They maintain that since the advent of routine PSA testing in the late 1980s, the mortality rate from prostate cancer of American men has been declining. Conversely, the rate of death due to prostate cancer continues to rise in countries where there is no routine PSA screening. They contend that this observation strongly supports the value of PSA screening. Their position is also supported by many experts. John Blanchard, president and CEO of the Prostate Cancer Research Foundation of Canada, states, "The PSA test is the most valuable tool we have for early detection of prostate cancer."¹⁰ Laurence Klotz, MD, Chief of Urology at Sunnybrook and Women's College Health Sciences Centre, Toronto, Ontario, affirms, "The PSA test provides physicians with an early indication of the cancer, and leads to other tests for the disease, including ultrasounds, biopsies, and the Gleason score, which evaluates how aggressive the cancer is."¹²

However, Hoffman disputes this position, arguing that proponents of PSA screenings are premature in their judgment, basing their reasoning on only 1 randomized Canadian study which yielded positive results.² He challenges the results of this study because they were reported after only 3 years, when prostate cancer is a "slow-growing cancer with a 5-year lead time."² Hoffman makes the case that, "just because screening rates have increased and mortality rates have decreased does not prove that mortality was lower among men who actually underwent screening."² He thinks that there may be other plausible explanations for the downward trend in mortality rates, such as better palliative treatment for advanced cancers. He adds that patients' lives are now being extended to the point that they die from other causes unrelated to their prostate cancer.

Recommendations Differ

Professional organizations differ in their recommendations for PSA screening of patients. The current recommendation from the American Cancer Society and the American Urologic Association is that all men age 50 years and older should have yearly PSA and DRE screening, and that African American men and those with a family history of prostate cancer should begin screening at age 45 years.^5,12 Conversely, the American Society of Internal Medicine, the American College of Physicians, and the US Preventative Services Task Force all recommend against routine screening.² According to Peter Carroll, MD, "From a functional standpoint, PSA works well as a test. Using PSA, you [do] improve prostate cancer detection. It's just that you may not want to detect it at an earlier stage."¹³ Carroll contends that with PSA, physicians cannot distinguish which patients have an aggressive cancer, and which ones have tumors which will never become clinically significant.¹³

In the New England Journal of Medicine (2004) and in the Journal of the American Medical Association (2005), 2 papers were published as a part of The Prostate Cancer Prevention Trial (PCPT) that examined the relationship between PSA and the risk of a diagnosis of prostate cancer if prostate biopsy were performed. In the first of these 2 papers, the investigators examined men who participated in the PCPT who had an end-of-study biopsy after 7 years.¹⁴ In contrast with previous studies, these researchers examined the risk of prostate cancer at PSA values below 4.0 ng/mL, a range that had previously been considered normal. As shown in Table 3, the risk of prostate cancer was elevated for all PSA concentrations, and increased with PSA values.¹⁵

The second study examined the relationship between prostate cancer and PSA concentration across the entire range of PSA concentrations, in order to help identify the PSA concentration at which a prostate biopsy should be performed. It was not limited to men with low PSA values.¹⁶ A diagnosis of prostate cancer was made for 1225 men (22%) from the placebo group of the PCPT who had a PSA test and a prostate biopsy.¹⁶

This study highlighted some important points: prostate cancer risk increased with the PSA value; prostate cancer was found even at low PSA concentrations; the higher the PSA level that is chosen to perform a biopsy, the greater the probability of prostate cancer, but the greater the risk of missing prostate cancer among men with lower PSA concentrations; and, conversely, choosing a lower PSA level as the cutoff to perform a biopsy identifies more men with prostate cancer, but also increases the number of unnecessary biopsies.¹⁵

To understand the usefulness of different PSA values as cutoffs for cancer screening, it is necessary to understand 2 key concepts: sensitivity and specificity.¹⁶ Sensitivity refers to the likelihood that, if the patient actually has cancer, the PSA test will be abnormal. A PSA level that is 90% sensitive will be elevated in 90% of men who have prostate cancer. If the goal of PSA screening is to detect 95% of men who have prostate cancer, then it will be necessary to use a lower PSA cutoff value for screening. The specificity of a PSA screening test is the likelihood that, for a patient without cancer, the test will be normal. A PSA value that has a 90% specificity will identify 10% of men who actually do not have cancer as "abnormal." The trade-off between the sensitivity and specificity of PSA screening is shown in Table 4.

The risk of prostate cancer decreases with lower levels of PSA. Therefore, to find more patients with cancer, the level at which PSA is called abnormal must be lowered. However, this will also cause more men who do not have cancer to have abnormal PSAs result and to undergo biopsies. The conclusion of these investigators was: "Biopsy-detected prostate cancer, including high-grade cancers, is not rare among men with PSA levels of 4.0 ng/mL or less, or levels generally thought to be in the normal range. There is no cut point of PSA with simultaneous high sensitivity and high specificity for monitoring healthy men for prostate cancer, but rather a continuum of prostate cancer risk at all values of PSA."¹⁶

In 2004, approximately 230 000 new cases of prostate cancer were diagnosed. However, of these cases, only 29 900 men died as a result of the disease.² This presents the "paradox" of PSA testing: "although prostate cancer is the second leading cause of cancer deaths in men, the almost 8:1 ratio in incidence to prostate cancer-specific mortality shows that most men do not die of their disease."³ Hoffman argues against routine PSA screening saying, "There is no conclusive evidence that screening is effective in reducing prostate cancer morbidity...but there is considerable evidence that screening and treatment can be harmful."² He asserts that positive results can cause patient anxiety and may prompt unnecessary biopsy and/or surgery, which carry their own risks. According to Hoffman, when microscopic cancers are detected, the expectation is to live for years, perhaps a lifetime, without complications.² However, complications from unnecessary treatment "will occur immediately and may persist for a lifetime."² Side effects from prostate biopsy include pain, blood in the semen (hematospermia), and urine (hematuria), in addition to fever. It also carries with it the same risks as any invasive procedure. Hoffman notes that 95% of men who have undergone radical prostatectomy have suffered urinary and sexual dysfunction.² Also, a substantial proportion (78%) of men have experienced sexual and bowel dysfunction after radiation therapy.

With a diagnosis of prostate cancer, physicians and patients must weigh certain factors when considering how to proceed. They must decide which is the lesser of 2 evils—withhold treatment and risk developing metastatic disease, or treat and risk possibly devastating side effects. Carroll emphasizes, "Physicians need to stress surveillance instead of aggressive treatment for patients whose prostate cancers are detected early."¹³ This may be a difficult message to accept for a nervous patient who has just received a diagnosis of cancer. Delaying treatment to observe whether the cancer will be aggressive is something most patients are not willing to do. According to Hoffman, "...men with early stage cancer are uncomfortable with watchful waiting."² Carroll urges, "We need to re-educate patients [about the benefits of surveillance versus treatment], and we need to do it before the biopsy."¹³

Size and Grade of the Tumor

There is also controversy surrounding the sensitivity of PSA screenings with regard to tumor size. Through the use of biopsy, tumor grade can be correlated with aggressiveness and can be charted on the Gleason scale. The Gleason score system ranks tumor aggressiveness on a scale between 2 and 10.¹ Tumor biopsies ranking between 2 and 4 designate low aggressiveness. Biopsy scores between 5 and 6 are considered mildly aggressive, those with a score of 7 are moderately aggressive, and scores of 8 to 10 are very aggressive.¹ However, tumor volume is more difficult to ascertain. Presently there is not an imaging study that accurately measures tumor volume.¹ Oottamasathien and Crawford have reported that tumors ranging from 0.5 to 1.9 cm³ generally produce enough PSA to be detected easily in screenings, and microscopic tumors will not be detected.¹ They assert that this proves that most cancers detected by PSA screening are large enough to be dangerous.¹

Stamey, a pioneer of PSA testing, thinks that much smaller tumors that pose little risk are being detected with current screening techniques, calling them "tumors which do not necessarily require aggressive treatment."¹⁷ He argues that subsequent treatment of these clinically insignificant tumors results in unnecessary harm to the patient, such as the aforementioned impotence, incontinence, and bowel disorders.¹⁷ Stamey states further that, "when we first started the PSA test, it showed a 40% to 60% relationship [specificity] with the large cancers. However, the surprising thing is that over the past 5 years it has fallen to 2%, presumably due to over screening in the United States."¹⁷

In a study spanning 20 years, Stamey examined 1317 surgically removed prostates and found that PSA screening during the first 10 years showed a fair correlation between PSA levels and volume of prostate tumors.¹⁷ However, during the latter 10 years, the study revealed that PSA correlated only with prostate size and that most men had surgery due to BPH alone. Stamey concluded, "Most of the prostates we remove need not be removed."¹⁷

Better Marker Needed

Although there is heated debate regarding the use of the PSA test, experts agree that a better screening tool is needed to distinguish between critical, aggressive cancers, and those that are slow-growing and may not become life-threatening. Unfortunately, at the time of diagnosis, men who need treatment for prostate cancer cannot be differentiated from men who do not. According to Troyer et al, "The most important challenge in prostate cancer today is the inability to predict the behavior of an individual tumor in an individual patient."¹⁸ They further state that, "the ultimate goal would be not only to have accurate biomarkers suitable for early diagnosis, but also biomarkers that identify men at greatest risk of developing aggressive disease."¹⁸

New Screening Techniques

New research is now underway which is yielding promising results in finding more accurate markers for distinguishing aggressive forms of prostate cancer from more slow-growing forms. Two of the major approaches are genomics and proteomics. With the advent of the mapping of the human genome in 2001, scientists are gaining a new understanding of the genes involved in prostate cancer. New DNA microarray technology is enabling researchers to examine thousands of genes at the same time.¹⁹ A group of researchers at the University of California, San Francisco, including Carroll, have identified several "chromosomal changes that correlate with progression and identified one that may predict progression independent of tumor stage and grade."²⁰ They have identified genes that are indicators of aggressive forms of cancer. Developing a serum marker for these genes may provide physicians with the ability to distinguish patients who need aggressive treatment from those who do not.

Mutation of 1 gene, EphB2 located on chromosome 1, was found during the African American Hereditary Prostate Cancer study.²¹ This study demonstrated the first gene mutation to be associated with familial prostate cancer in African American men. The study implicated EphB2 as a prostate cancer tumor-suppressor gene.²² The investigators concluded that "the K1019X mutation in the EphB2 gene differs in frequency between African American men and European American men, is associated with increased risk for prostate cancer in African American men with a positive family history, and may be an important genetic risk factor for prostate cancer in African American men."²²

In February 2006, the National Institutes of Health launched a new initiative to identify genetic risk factors for breast and prostate cancer.²² Cancer Genetic Markers of Susceptibility (CGEMS) is a 3-year initiative, funded for $14 million, which will conduct scans of the entire human genome (genotyping) to identify common, inherited gene mutations that increase the risks for breast and prostate cancer. "CGEMS is among the first large whole genome scanning projects in cancer, and we are hopeful that its results will provide promising new insights into understanding genetic risk and common cancers, such as breast and prostate cancer," stated Stephen J. Chanock, MD, director of the National Cancer Institute's Advanced Technology Center Core Genotyping Facility.²²

Many researchers are also using proteomics to study prostate cancer. Although genes are the blueprint for cellular function, proteins are the building blocks to carry out their functions. Proteomics is the study of the composition, structure, function, and interactions of the proteins directing the activities of each living cell.²³ Each different type of cell creates its own set of proteins, giving it a unique proteome. Once normal cell proteomes have been established, changes that may indicate malignancy can be recognized. Researchers are currently working to establish normal prostate cell proteomes, thus markers can be identified in prostate cancer cells, and differentiate between various types and stages of the disease. One protein already identified is known as EZH2.¹⁹ According to the American Cancer Society, this protein "appears more often in advanced prostate cancer than in those at early stage."¹⁹ The experts at the Lineberger Comprehensive Cancer Center at the University of North Carolina, Chapel Hill, state that proteomics can discriminate between men with higher PSA levels caused by BPH and men with prostate cancer even when the PSA levels are in the uncertain "gray zone."²⁴ They add, "We are currently expanding the testing set to determine the reliability of this new technology to decrease unnecessary prostate biopsies without compromising the detection of curable [prostate cancer]."²⁴ The proteome is only beginning to be understood, and certainly, a vast amount of work is yet to be done. Troyer et al conclude, "This huge task is only beginning and requires the combined expertise of molecular epidemiologists, oncologists, surgeons, pathologists, and basic scientists."¹⁸

Conclusions

According to Stamey, "[Prostate cancer is] a cancer we [men] all get if we live long enough."²⁵ Although currently the PSA test is the screening tool of choice to detect prostate cancer, it must be considered unreliable at best. Efforts to improve the PSA test's accuracy are ongoing, and scientists should continue to work to find an alternate tumor marker that will differentiate aggressiveness. Certainly, PSA testing can be useful if applied properly; by combining the results with other available screenings. However, even then, it must be used with caution because of the slow-growing nature of the disease.

Physicians and patients need to be well informed regarding the facts about prostate cancer, and at diagnosis they should carefully evaluate whether to treat the disease aggressively or simply to monitor its progress. A positive PSA does not necessarily deliver a death sentence to the patient, and a well-informed patient will not be frightened into undergoing a treatment that may be unnecessary.

Many studies are being conducted in the search for a better tumor marker. Until this is a fact and not a possibility, perhaps it is best to continue screening with the tools at hand using caution born of experience with a less than perfect screening tool—the PSA test.

References

1. Oottamasathien S, Crawford ED. Should routine screening for prostate-specific antigen be recommended? JAMA. 2003;163:661-663.

2. Hoffman RM. An argument against routine prostate cancer screening. JAMA. 2003;163:666.

3. What are the key statistics about prostate cancer? American Cancer Society Web site. Available at: http://www.cancer.org/docroot/CRI/content/CRI_2_4_1X_What_are_the_key_statistics_
for_prostate_cancer_36.asp. Accessed March 27, 2005.

4. Screening for prostate cancer. WebMDHealth Web site. Available at: http://my.webmd.com/content/article/5/1680_50803.htm. Accessed: March 31, 2005.

5. Can prostate cancer be found early? American Cancer Society Web site. Available at: http://www.cancer.org/docroot/CRI/content/CRI_2_4_3X_
Can_prostate_cancer_be_found_early_36.asp. Accessed March 27, 2005.

6. PSA: blood test gives urologists a new way to detect prostate cancer. His and Her Health Web site. Available at: http://www.hisandherhealth.com/aua2002/5.html. Accessed March 30, 2005.

7. Prostate-specific antigen (PSA). Available at: http://www.webmd.com/hw/mens_conditions/hw5522.asp. Accessed August 13, 2006.

8. TC-Cancer Web site. Tumor markers: AFP, HCG, CA-125. Available at: http://www.tc-cancer.com/tumormarkers.html. Accessed October 6, 2006.

9. Henderson RJ, Eastham JA, Culkin DJ, et al. Prostate-specific antigen (PSA) and PSA density: racial differences in men without prostate cancer. J Natl Cancer Inst. 1997;89:134-138.

10. Koliakos G, Chatzivasiliou D, Dimopoulos T, et al. The significance of PSA/IGF-1 ratio in differentiating benign prostate hyperplasia from prostate cancer. Dis Markers. 2000;16:143-146.

11. The Doctor's Doctor Web site. Prostate-specific antigen. Available at: http://www.thedoctorsdoctor.com/labtests/prostate_specific_antigen.htm. Accessed October 6, 2006.

12. Do not abandon PSA test, organization warns. Urology Times. 2004;32: l6-20. Available at: http://www.findarticles.com/p/articles/mi_hb334/is_200411/ai_n13364089. Accessed April 11, 2005.

13. Mitka M. Is PSA testing still useful? JAMA. 2004;292:2326-2327.

14. 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.

15. Southwest Oncology Group. PSA and prostate cancer. Available at: http://www.swog.org/Visitors/Download/Clinical%20Trials/PCPT_PSA_Summary.pdf. Accessed August 15, 2006.

16. Thompson IM, Ankerst DP, Chi C, et al. Operating characteristics of prostate-specific antigen in men with an initial PSA level of 3.0 ng/mL or lower. JAMA. 2005;294:66-70.

17. Stamey TA. The era of serum prostate specific antigen as a marker for biopsy of the prostate and detecting prostate cancer is now over in the USA. BJU Int. 2004;94:963-964.

18. Troyer DA, Mubiru J, Leach RJ, et al. Promise and challenge: markers of prostate cancer detection, diagnosis, and prognosis. Dis Markers. 2004;20:117-128.

19. What's new in prostate cancer research and treatment? American Cancer Society Web site. Available at: http://www.cancer.org/docroot/CRI/content/CRI_2_4_6X_Whats_New_in_Prostate_Cancer_
Research_and_Treatment_36.asp. Accessed March 20, 2005.

20. University of California, San Francisco. Genomics of prostate cancer. Available at: http://cancer.ucsf.edu/prostate_spore/prjct_cc-pc.php. Accessed August 15, 2006.

21. Kittles RA, Baffoe-Bonnie AB, Moses TY, et al. A common nonsense mutation in EphB2 is associated with prostate cancer risk in African American men with a positive family history. J Med Gen. 2006;43:507-511.

22. NCI launches new initiative to identify genetic risk factors for breast and prostate cancer. NIH News. February 13, 2006. Available at: http://www.nih.gov/news/pr/feb2006/nci-13.htm. Accessed August 13, 2006.

23. What is proteomics? Available at: http://www.learner.org/channel/courses/biology/textbook/proteo/proteo_1.html. Accessed April 12, 2005.

24. Ornstein DK, Rayford W, Fusaro VA, et al. Serum proteomic profiling can discriminate prostate cancer from benign prostates in men with total prostate specific antigen levels between 2.5 and 15.0 ng/mL. J Urol. 2004;172:1302-1305.

25. Early promoters of PSA screening do a turnabout. Center for Medical Consumers Web site. Available at: