ABSTRACT: Although the widespread use of prostate-specific antigen (PSA) testing has led to an increase in the number of cancers detected, controversies about the benefits of screening persist. No conclusive evidence has yet emerged that PSA screening reduces the mortality associated with prostate cancer. Thus, mass screening is not universally endorsed. The American Urologic Association and the American Cancer Society recommend that digital rectal examination and PSA testing be offered annually to men 50 years and older with an estimated life expec- tancy of 10 years or more. High-risk patients (those with a positive family history or those of African American descent) are advised to begin screening at age 45. The decision to screen is based on the patient's preference following a thorough discussion of the benefits and limitations of PSA testing. Refer to a urologist any patient with a PSA greater than 4.0 ng/mL. Also, be alert for high PSA velocity changes in patients undergoing annual screening, and refer those with a PSA velocity of more than 0.75 ng/mL/y.
A 57-year-old white man with no urologic symptoms and no family history of prostate cancer requests a prostate- specific antigen (PSA) test. What will you tell him?
Prostate cancer detection and treatment have been profoundly affected by the advent of PSA testing. Nonetheless, the specificity and sensitivity of the assay are imperfect, which has resulted in disagreement about the test's accuracy in the diagnosis of cancer and the overall treatment benefit of mass PSA screening. Data are not yet available from randomized trials to determine whether early detection is beneficial or harmful or has no effect. Consequently, the optimal strategy for early detection with PSA testing remains unclear.
Current clinical guidelines and the evolving refinement of PSA assays have addressed some of these concerns. However, further standardization of screening and treatment guidelines is needed.
In this article, we review the principal risks and benefits of screening and highlight important areas of controversy.
THE IMPACT OF PSA TESTING
PSA--a serine protease produced by the epithelial cells of the prostate gland--is secreted into the prostatic ducts and aids in the liquefaction of semen.1 Early investigation revealed that PSA was specific to the prostate gland, which suggested that this antigen might be helpful in the diagnosis and/or treatment of prostate cancer.2 In 1986, the FDA approved PSA testing for use in the monitoring of prostate cancer recurrence following prostatectomy.3 PSA testing was approved by the FDA for prostate cancer screening in 1994 and has had a substantial impact on the detection and treatment of prostate cancer.3
The principal result of PSA testing has been the increasing number of cancers detected annually.4 Not surprisingly, this trend has paralleled an increase in prostate cancer cases diagnosed as a result of elevated PSA levels. Data reported approximately 10 years after the introduction of PSA testing revealed that prostate cancer detection as a result of PSA elevation had become as common as detection through an abnormal digital rectal examination (DRE) and was far more common than incidental discovery following transurethral resection of the prostate in the treatment of benign prostatic hyperplasia (BPH).5 As a result of this trend, cancer confined to the prostate is being diagnosed in a larger number of patients, with a corresponding reduction in the incidence of lymph node metastases detected during radical prostatectomy.5,6
Despite these trends, several concerns remain. Earlier diagnosis of organ-confined disease has been accompanied by advances in the surgical techniques used in radical prostatectomy, which should, theoretically, result in improved cancer-specific survival. However, Soh and colleagues5 found no change in pathologic stage or prognosis for patients treated during a 12-year period before and after the introduction of PSA screening. In addition, randomized controlled trials have not conclusively demonstrated that PSA screening reduces the mortality associated with prostate cancer.7
Usefulness of PSA screening. Although the detection of increasing numbers of prostate cancers through PSA screening might seem an important accomplishment, autopsy data show that 80% of men in their 70s have prostate cancer.8 Significant, potentially life-threatening disease develops in only a small percentage of men.
Data that were obtained around the time PSA screening was introduced showed a correlation between PSA level and tumor volume. Large tumor volume, in turn, suggested more aggressive disease.9 However, recent data have demonstrated that the relationship between PSA level and tumor volume has disappeared, which has led some authors to conclude that the PSA screening era is "over."9
Accordingly, the concept of clinically insignificant prostate cancer has been developed, and research has increasingly focused on the use of pathologic variables (such as Gleason grade and tumor volume) to better define cancers that may be more indolent.10 Despite these steps, it is still not possible to identify aggressive tumors selectively. For this reason, controversy remains centered around the utility of mass screening programs and the potential for overdiagnosis.
Benefits of intervention. Controversy surrounding PSA screening is also related to the natural course of prostate cancer and the variable benefit of intervention in patients in whom the disease is diagnosed. A seminal report by Pound and associates11 found a median actuarial time of 8 years from the recurrence of PSA elevation after radical prostatectomy to the development of metastatic disease. An additional 5-year actuarial time from metastatic disease to death was reported in these patients.
Based on these data, it is evident that aggressive intervention is warranted only in those patients with an appropriate life expectancy, because a significant number of patients--even those with untreated prostate cancer--will die of other causes. Further, because of the slow rate of disease progression in most patients, a lengthy window of opportunity may exist for treatment. Therefore, immediate intervention is not always mandatory, especially for patients in whom life expectancy may be limited.Based on these considerations, some authors advocate watchful waiting as an acceptable treatment approach in patients with a limited life expectancy or in those with less aggressive tumors.12,13However, selection of patients for whom expectant management is appropriate remains difficult.
Sensitivity of PSA. Much attention has been paid to the lack of specificity of PSA for prostate cancer; however, controversy also exists about the assay's sensitivity. Although the commonly used PSA threshold of 4.0 ng/mL results in reasonable sensitivity, some experts believe that a lower PSA threshold is needed to increase cancer detection.14 However, lowering the PSA threshold for prostate biopsy would probably heighten existing concerns about overdiagnosis of indolent disease.10 These concerns have led researchers to attempt to identify cancers with a lower potential for progression and death.15
Although PSA testing has undoubtedly enhanced the ability to detect prostate cancer, the lack of specificity has resulted in considerable controversy about such questions as standardization of reference ranges and the appropriate indications for prostate biopsy. The lack of consensus is most evident in recent clinical guidelines for PSA screening, which vary considerably between organizations, and in the differing approaches to routine screening observed in primary care practices.16
The latest American Urologic Association and American Cancer Society guidelines recommend that prostate cancer screening be offered annually to men 50 years and older with an estimated life expectancy of 10 years or more.17,18 High-risk patients (those with a positive family history or those of African American descent) are advised to begin screening at age 45 years. According to these groups, prostate cancer screening should include both PSA measurement and DRE. Other organizations, such as the US Preventive Services Task Force, believe that there is insufficient evidence to support prostate cancer screening.3 The CDC does not recommend routine screening because of the lack of consensus about the benefits of PSA screening with regard to mortality.19
Despite this controversy, most clinical guidelines--including those of the American College of PhysiciansAmerican Society of Internal Medicine and the American Academy of Family Physicians--suggest that screening decisions be individualized based on patient preference, following an informed discussion that addresses the benefits and limitations of screening (Box).
A discussion of these issues is equally important in patients with a life expectancy of less than 10 years. Although we recommend against routine PSA screening in this group,we find that patients are often uncomfortable with the decision to forgo PSA screening, largely because of widespread publicity about this test. In these cases, a reasonable alternative to annual PSA screening is a one-time PSA measurement, with follow-up based on the initial result.
The same concerns are also important in deciding when to discontinue annual PSA testing in patients of advancing age who have undergone screening for many years.We suggest to our patients who choose to forgo PSA screening that we would like to continue to perform an annual DRE, which may detect, and allow for treatment of, high-volume or more aggressive prostate cancers.
PSA REFERENCE RANGE
Based on early investigations that demonstrated an increased risk of prostate cancer in men with a PSA level greater than 4.0 ng/mL, this level is typically used as the threshold to recommend prostate biopsy.20 Nevertheless, the sensitivity of this level is limited. Table 1 illustrates the associated probability of cancer detection by subsequent biopsy stratified by various PSA ranges. More than 30% of men with organ-confined prostate cancer may have a PSA level of less than 4.0 ng/mL.21,22
Based on recent investigation, some authors recommend that a threshold of 2.5 ng/mL be used to recommend prostate biopsy.14 Related research suggests that a higher percentage of patients may receive a diagnosis of organ-confined cancer with use of this threshold and that this adjustment will not result in overdetection of clinically insignificant prostate cancer.3 However, other authors remain concerned that lowering PSA reference ranges will result in the diagnosis of more cases of clinically indolent disease.10
Additional questions about the appropriate PSA reference range are related to PSA variations that may result from a patient's age or race (Table 2). Thus, some investigators have suggested that age-specific and race-specific reference ranges may be helpful.23,24 However, evidence indicates that a significant number of cancers will remain undetected even with the use of modified reference ranges.25
Because no data have conclusively demonstrated that modified reference ranges improve outcomes, we continue to recommend prostate biopsy in patients with a PSA level greater than 4.0 ng/mL. In patients at increased risk and in those younger than 55 years with PSA values of 2.5 to 4.0 ng/mL, we may use supplemental PSA assays (detailed below) to further help stratify risk, but we would more commonly recommend biopsy.
OPTIMIZING PSA ACCURACY
Serum PSA elevation is seen in a number of benign processes, including BPH, urinary retention, recent ejaculation, urethral catheterization, and prostatitis.3 Although some experts recommend that PSA measurement be avoided within 1 week of DRE, evidence suggests that DRE has minimal or no effect on serum PSA concentration in the range below 20 ng/mL.26 We most commonly measure PSA before the DRE, although this is not mandatory. Other evidence has shown that elevated PSA levels are associated with cycling; this has led some authors to recommend against PSA testing within 1 week of this activity.27A newer study did not find a connection between cycling and PSA levels.28 Nonetheless,we advise patients to avoid cycling and sexual activity during the week before PSA measurement.
We routinely obtain a urinalysis as part of the annual physical examination, which serves to exclude any infectious cause for PSA elevation. If a patient presents with symptoms that suggest a urinary tract infection, a complete urinalysis and microscopic examination are warranted. If you suspect that the PSA level is spuriously elevated, it is reasonable to repeat the measurement after an appropriate interval (eg, 4 to 6 weeks).
OTHER PSA ASSAYS
A number of assays are commonly used by urologists to help distinguish between benign and malignant causes of PSA elevation.
PSA velocity. This is defined as the rate of rise in PSA level over time. Interest in PSA velocity was based on the premise that the rapidity of cell division associated with prostate cancer might be greater than that associated with BPH. An increase in the serum PSA of 0.75 ng/mL/y or more suggests prostate cancer.29 Using this threshold, PSA velocity has been associated with a specificity of 95% in men with PSA values below 10 ng/mL.20
PSA density (PSAD). This is defined as the serum PSA divided by prostate volume. PSAD measurement was introduced in an attempt to control for PSA elevation resulting from BPH. A PSAD greater than 0.15 ng/mL/cc is reported to indicate cancer, although some experts believe that this cutoff results in many missed cancers.20,30 Further, measurement of PSAD requires transrectal ultrasonography, which is associated with additional cost and discomfort.
Total, free, and complexed PSA. Serum PSA is present in 2 general forms--free PSA (fPSA) and complexed PSA (cPSA). Most of the PSA that enters the systemic circulation becomes complexed to protease inhibitors, the most common of which is alpha-1-antichymotrypsin. The relative percentage of fPSA to total PSA (tPSA) is known as %fPSA. Investigation has shown that men with prostate cancer are more likely to have a lower %fPSA than those who do not have prostate cancer.20 Currently, there is no consensus regarding the %fPSA threshold at which biopsy is recommended; %fPSA is used instead to further stratify the risk of patients with an elevated PSA value. For example, a %fPSA greater than 20% is associated with an 8% risk that cancer will be detected on subsequent biopsy, whereas a %fPSA of less than 10% is associated with cancer in 46% of patients.31
We find that a %fPSA assay is most useful when it is unclear whether a biopsy is necessary. In such cases, we recommend biopsy for patients with a %fPSA of less than 10%. If the %fPSA is greater than 20%, we usually wait and repeat the test in 6 to 12 months.
Usefulness of assays. A true determination of the utility of these assays remains difficult. We believe that these tests may be used to complement routine tPSA measurement. They are most useful in the PSA "gray zone"--that is, the range between 4 and 10 ng/mL. They are also helpful in guiding the decision to perform repeated prostate biopsy in patients in whom previous biopsy yielded no evidence of cancer but who consistently have elevated and/or increasing PSA levels.
Accordingly, in our practice, patients with a tPSA of more than 4 ng/mL and a life expectancy of at least 10 years are routinely offered prostate biopsy. In addition, the finding of a palpably abnormal prostate that suggests malignancy generally warrants biopsy, irrespective of PSA level. Patients who have no evidence of malignancy on prostate biopsy are then monitored closely, and a repeated tPSA is performed annually or semiannually. In these patients, we often follow tPSA with a combination of PSA velocity, %fPSA, and/or cPSA assays. Although some literature suggests that the %fPSA may serve as a more specific marker of cancer than does PSAD or PSA velocity, no clear evidence suggests the superiority of any single assay.32
Controversy exists about the benefit of supplemental assays in patients with PSA values less than 4.0 ng/mL. For example, whereas some investigators believe that PSA velocity is useful in patients with a PSA level in this range, others think that assay specificity is more limiting.3,20,33 Despite this controversy, we believe that a high PSA velocity, even with a tPSA of less than 4.0 ng/mL, is cause for concern, and that closer follow-up and/or additional PSA assays may be useful in this setting.
The power of any PSA assay to detect prostate cancer is limited. For this reason, we suggest that any patient with a PSA greater than 4.0 ng/mL be referred to a urologist for further counseling and/or intervention. Further, we recommend that you also be alert for high PSA velocity changes in patients undergoing annual screening and refer those with a PSA velocity of more than 0.75 ng/mL/y.
Narayan P. Neoplasms of the prostate gland. In: Tanagho EA, McAninch JW, eds.
Smith's General Urology.
14th ed. Norwalk, Conn: Appleton & Lange; 1995.
Ablin RJ, Bronson P, Soanes WA, Witebsky EO. Tissue- and species-specific antigens of normal human prostatic tissue.
Han M, Gann PH, Catalona WJ. Prostate-specific antigen and screening for prostate cancer.
Med Clin North Am.
Parker SL, Tong T, Bolden S, Wingo PA. Cancer statistics, 1997.
CA Cancer J Clin.
Soh S, Kattan MW, Berkman S, et al. Has there been a recent shift in the pathological features and prognosis of patients treated with radical prostatectomy?
Catalona WJ, Smith DS, Ratliff TL, Basler JW. Detection of organ-confined prostate cancer is increased through prostate-specific antigen-based screening.
Chan EC, Vernon SW, Haynes MC, et al. Physician perspectives on the importance of facts men ought to know about prostate-specific antigen testing.
J Gen Intern Med.
Sakr WA, Grignon DJ, Haas GP, et al. Age and racial distribution of prostatic intraepithelial neoplasia.
Stamey TA, Caldwell M, McNeal JE, et al. The prostate specific antigen era in the United States is over for prostate cancer: what happened in the last 20 years?
2004;172(4 pt 1):1297-1301.
Hernandez J, Thompson IM. Prostate-specific antigen: a review of the validation of the most commonly used cancer biomarker.
Pound CR, Partin AW, Eisenberger MA, et al. Natural history of progression after PSA elevation following radical prostatectomy.
Chodak GW, Thisted RA, Gerber GS, et al. Results of conservative management of clinically localized prostate cancer.
N Engl J Med.
Albertsen PC, Fryback DG, Storer BE, et al. Long-term survival among men with conservatively treated localized prostate cancer.
Punglia RS, D'Amico AV, Catalona WJ, et al. Effect of verification bias on screening for prostate cancer by measurement of prostate-specific antigen.
N Engl J Med.
Bastian PJ, Mangold LA, Epstein JI, Partin AW. Characteristics of insignificant clinical T1c prostate tumors. A contemporary analysis.
Cooper CP, Merritt TL, Ross LE, et al. To screen or not to screen, when clinical guidelines disagree: primary care physicians' use of the PSA test.
American Urologic Association. Prostate-specific antigen (PSA) best practice policy.
2000; 14:267-272, 277-280.
Smith RA, von Eschenbach AC, Wender R, et al. American Cancer Society guidelines for the early detection of cancer: update of early detection guidelines for prostate, colorectal, and endometrial cancers.
CA Cancer J Clin.
Centers for Disease Control and Prevention. Prostate cancer: the public health perspective. 2003. Available at: www.cdc.gov. Accessed February 17, 2005.
Gretzer MB, Partin AW. PSA markers in prostate cancer detection.
Urol Clin North Am.
Schroder FH, van der Cruijsen-Koeter I, de Koning HJ, et al. Prostate cancer detection at low prostate specific antigen.
Thompson IM, Pauler DK, Goodman PJ, et al. Prevalence of prostate cancer among men with a prostate-specific antigen level N Engl J Med. 2004;350:2239-2246.
Oesterling JE, Jacobsen SJ, Cooner WH. The use of age-specific reference ranges for serum prostate specific antigen in men 60 years old or older.
Moul JW, Sesterhenn IA, Connelly RR, et al. Prostate-specific antigen values at the time of prostate cancer diagnosis in African-American men.
Catalona WJ, Southwick PC, Slawin KM, et al. Comparison of percent free PSA, PSA density, and age-specific PSA cutoffs for prostate cancer detection and staging.
Crawford ED, Schultz MJ, Clejan S, et al. The effect of digital rectal examination on prostate-specific antigen levels.
Weston R, Parr N. New NHS guidelines for PSA testing in primary care.
Lippi G, Corgnati A, Salvagno G, et al. Total and free PSA serum concentrations are not influenced by extensive physical exercise and bicycle riding.
Int J Sports Med.
Carter HB, Pearson JD, Metter EJ, et al. Longitudinal evaluation of prostate-specific antigen levels in men with and without prostate disease.
Catalona WJ, Richie JP, de Kernion JB, et al. Comparison of prostate specific antigen concentration versus prostate specific antigen density in the early detection of prostate cancer: receiver operating characteristic curves.
1994;152(6 pt 1): 2031-2036.
Catalona WJ, Partin AW, Finlay JA, et al. Use of the percentage of free prostate-specific antigen to identify men at high risk of prostate cancer when PSA levels are 2.51 to 4 ng/mL and digital rectal ex-amination is not specific for prostate cancer: an al-ternative model.
Parsons JK, Partin AW. Applying complexed prostate-specific antigen to clinical practice.
Djavan B, Zlotta A, Kratzik C, et al. PSA, PSA density of transition zone, free/total PSA ratio, and PSA velocity for early detection of prostate cancer in men with serum PSA 2.5 to 4.0 ng/mL.
Moul JW. The prospects for targeted screening for prostate cancer.
Prospectives Contemp Issues Managing Prostatic Dis.
Oesterling JE, Jacobsen SJ, Chute CG, et al. Serum prostate-specific antigen in a community-based population of healthy men. Establishment of age-specific reference ranges.
Presti JC Jr, Hovey R, Bhargava V, et al. Prospective evaluation of prostate specific antigen and prostate specific antigen density in the detection of carcinoma of the prostate: ethnic variations.