Navigation

Prostate Cancer Staging

Prostate cancer is second only to lung cancer as a cause of cancer-related mortality, and early diagnosis is essential for improving survival. The three tests commonly used to detect prostate cancer are the digital rectal exam (DRE), transurethral ultrasound (TRUS), and monitoring serum prostate specific antigen (PSA) levels.

The American Cancer Society reports that in 1997, prostate cancer was diagnosed in over 200,000 men, and caused an estimated 40,000 deaths.[1] Today, prostate cancer is the most commonly diagnosed type of cancer among men, and it also accounts for over one-third of newly diagnosed cancers in American men and 13% of cancer-associated deaths.[1] Prostate cancer is second only to lung cancer as a cause of cancer-related mortality.[1]

Prostate cancer affects mostly older men; diagnosis before age 40 years is rare, and the median age at diagnosis is 70 years.[2] Data from a study sponsored by the National Cancer Institute involving patients from nine U.S. cancer centers indicated that the incidence of prostate cancer increased 6.4% per year on average among Caucasian males ages 59 to 79 years.[2] The greatest increase (7.7%) was in men ages 60 to 69 years. Results from this study also revealed a substantial increase in prostate cancer (from 276 per 100,000 men to 383 per 100,000 men) between 1983 and 1989. The spurt is probably due to the increased use of screening tests, including measurement of serum prostate specific antigen (PSA) and the digital rectal exam (DRE); an increased awareness of the disease in the population; and the fact that more men are living longer.

Expression of prostate cancer varies, and it is often undetected. Some autopsy studies have shown that nearly 30% of men older than age 50 years have evidence of malignancy in the prostate gland, with no clinical evidence of the disease.[2] When detected at autopsy, the latent cancer may be only a microscopic lesion that never became symptomatic during a person's lifetime. On the other hand, a clinically significant tumor may be found that could decrease the patient's life span if treatment had not been initiated at once. The incidence of autopsy-detected prostate cancer, of whatever nature, increases to over 40% in men older than age 75 years, and the incidence approaches 100% in men ages 90 years and older.[2] For a 50-year-old man, the estimated lifetime risk of prostate cancer is 42%, the risk of clinical disease is 9.5%, and the risk of death from the disease is 2.9%.[2]

Etiology, signs, and symptoms of prostate cancer

The cause of prostate cancer is unknown. Several risk factors have been suggested, including too little dietary fiber, too much dietary fat, and too much (or too little) Vitamin A.[3] Heredity seems to be important; the likelihood of prostate cancer in a man with a father or brother who has the disease is twice as great (16% vs. 8%) when compared with a man with a no family history of the disease.[3] In the United States, prostate cancer is more common among African-Americans than Caucasians and is least common within the Asian-American population.[3] However, Japanese men experience an increased incidence if they move to the United States.[3] The incidence of prostate cancer also seems to be greater among printers, painters, rubber workers, loggers, ship fitters, farmers, and drag and chemical workers.[3] However, other than family history, most of the data on risk factors are conflicting or inconclusive.[3]

Most patients are asymptomatic at the time of diagnosis. If symptoms are present, they may be confused with those that accompany benign prostatic hyperplasia (BPH). These symptoms include pain with urination, increased frequency of urination, hematuria, a decline in the force of the urinary stream, dribbling, incomplete emptying of the bladder, and nocturia.[4] Consequently, screening tests that can identify men with possible prostate cancer are essential. Three tests commonly used for this purpose include DRE, transurethral ultrasound (TRUS), and serum PSA levels.

Characteristics of PSA

Prostate-specific antigen is a glycoprotein weighing approximately 33 to 34 kDa. Structurally, PSA is similar to other proteases, with disulfide bonds (five for PSA) that help produce a three-dimensional configuration for the molecule's active site. In several reports, the half-life for PSA varies from 1.9 to 4.6 days.[5]

As a member of the kallikrein family of neutral proteases that includes kallikrein I, PSA has 50-80% homology with other proteases.[5-7] PSA is one of several proteolytic enzymes that causes liquefaction of semen immediately following ejaculation. PSA is produced by the epithelial cells of the prostate, and is present in seminal fluid, serum, and urine. Recent studies have also found PSA in periurethral tissue and parotid glands.[8] (See Table 1.)

In serum, PSA exists in several free and complexed forms, with at least four isotypes [ILLUSTRATION OMITTED]. In healthy men, approximately 80% of PSA is complexed or bound to one of three protease inhibitors, primarily with [[Alpha].sub.1]-antichymotrypsin (ACT), but also with [[Alpha].sub.1]-antitrypsin ([A.sub.1]A), or [[Alpha].sub.2]-macroglobulin ([A.sub.2]M). Approximately 20% of PSA circulates as free (f)PSA.

Measurement of serum PSA

Various commercial procedures detect total PSA by sandwiching PSA between two antibodies. The first antibody is directed at and captures one site (epitope) on the PSA molecule in the serum sample. A second antibody molecule, tagged with an enzyme or other marker, binds to a second site on the PSA molecule to complete the sandwich. Signal measurement for this type of assay is directly proportional to the concentration of PSA in the sample.

Unresolved analytical issues[9-12] are inherent in this approach for detecting PSA:

* Different commercial assays use different mono- or polyclonal antibodies, which can yield different results (see Table 2).

* Different sources of calibrators can be mixtures of complexed and free PSA, which cause variations between methods.

* More than one form of PSA is present in the serum. In the range from approximately 4 ng/mL to 10 ng/mL, it is especially important for the assay to measure equally any form of complexed and fPSA. Such an assay could be termed equimolar, indicating that the assay antibodies will combine with a site on the PSA molecule, regardless of whether the PSA was originally complexed in the serum.

* Results from a recent survey suggest that equimolar binding does not occur in all assays (see Table 2).[9-11] Striking differences exist between methods; variations of more than 50% can occur within an assay. This latter point must be considered when monitoring a patient who is either being treated for prostate cancer or being followed, but not yet treated.

All of these analytical issues must be considered when interpreting data from different laboratories and when interpreting derivative assays such as PSA density and PSA velocity (described below). Most authors consider PSA to be the most sensitive marker of prostate cancer, and most laboratories have replaced the assay of prostate acid phosphatase (PAP) with PSA to screen for and monitor prostate cancer. However, PSA is not specific for prostate cancer and may be increased by many prostatic diseases, manipulations, or other conditions (see Table 1).

Detection of prostate cancer: Which tests to use?

Of the three commonly used tests for prostate cancer (DRE, TRUS, and serum PSA levels), the positive predictive values (PPV) are 22% to 36% for DRE, 15% to 41% for TRUS, and 22% to 35% for PSA levels greater than 4.0 ng/mL, but 65% to 67% for PSA levels greater than 10.0 ng/mL.[13-15] When both the DRE and the TRUS are abnormal, the PPV is 37% to 61%.[13-15] Combining the PSA and TRUS yields a PPV of 33% to 52%.[13-15]

When all three tests are abnormal, use of this combination yields the best PPV (62% to 74%).[13-15] However, the data for all three tests are only somewhat better than using just the DRE and PSA (69% for levels greater than 10.0 ng/mL), and the use of all three tests markedly increases the cost of screening.[2] Therefore, if the PPV could be increased for results obtained when using just the DRE and PSA (or some other cost-effective method), use of the TRUS could be minimized while improving patient care and reducing costs. It is recommended that DRE and PSA be used as initial screening tests.[4,13-15]

Attempts to improve the diagnostic value of PSA results Age-specific reference intervals.

Normally, as a man ages the prostate enlarges and produces more PSA. Therefore, age-specific normal ranges (see Table 3) have been suggested as a more useful aid to detect prostate cancer than the 0 to 4 ng/mL previously considered normal, regardless of a man's age.[16]

Use of age-specific normal ranges has several advantages:

* It can account for the continued enlargement of the prostate.

* It provides more sensitivity for detecting prostate cancer in younger men. * It provides greater specificity in older men.

* It may eliminate the need for measuring the volume of the gland to calculate the PSA density.[9-12]

It should be noted that results supporting the use of age-specific reference values were from a study that evaluated 471 white males in a single community, while other investigators did not confirm this.[17-20] For example, Catalona[21] found that using a cut-off of 4.0 ng/mL for all men was superior to age-specific values. Results from yet another study support using age-specific reference values for patients under age 60 years, but suggest that a high cut-off value of 4.0 ng/mL be used for patients older than age 60 years, to detect as many as 60% of older men with prostate cancer.[8]

PSA density. As the prostate gland enlarges with age, its volume increases from an average of 25 cc for men between ages 40 and 49 years, to an average of 40 cc for men between ages 70 and 79 years.[13] This observation, coupled with Oesterling's observation that the reference interval for PSA seems to increase with age, led to the idea that PSA density (PSA/size of gland) may help detect prostate cancer.[19,22-25]

For example, using a cut-off of 0.15 ng/mL for PSA density, Seaman et al.[26] compared PSA density with PSA results and found that the PSA density had a higher prostate cancer PPV than did PSA. Since then, further data did not substantiate the value of PSA density.[27] Pannek and Partin[8] concluded that PSA density is only valuable for identifying patients with intermediate PSA levels (4.0 to 10.0 ng/mL), and that these patients have a low risk for prostate cancer. Beduschi and Oesterling[28] suggested performing an annual DRE and PSA. If the PSA values were between 4 and 10 ng/mL, then the PSA density was calculated.

PSA velocity. The fact that PSA values increase with age and that a growing tumor often produces more than a normal amount of PSA for a given age led to the proposal that the rate of change (velocity) for PSA might be diagnostic. Carter et al.[29] found that an increase in the PSA of 0.75 ng/mL per year was a better discriminator of BPH, prostate cancer, and normal conditions than a single PSA value. While the sensitivity of PSA velocity for detecting prostate cancer remained at approximately 75%, the specificity improved from 60% to 90% for men with BPH versus normal. Other researchers substantiated this cut-off.[30,31]

Unfortunately, two limitations may minimize the use of PSA velocity. First, before PSA velocity can be used, three or more values must be obtained within a period of 24 months.[16] These values may be affected by intra-individual variation, intra-assay variation, and variation between assays (see Table 2) if a patient relocates to another healthcare system. Second, an initial abnormal value would probably prompt at least one biopsy that might then be performed earlier than warranted.

fPSA and fPSA/total PSA. As mentioned previously, PSA exists in several forms in the serum. In the early 1990s, studies[5,8] indicated that the predominant form of PSA in serum is a complex with ACT, and that this complex is often increased in patients with prostate cancer. Following this observation, several studies examined the utility of this difference as a tool to detect prostate cancer and to differentiate it from BPH.[32-37]

Table 2

PSA survey results

Method             n(*)       Sample 1       Sample 2      Sample 3
Abbott Axsym       1080          32             1.1          11.5
                                5-39          0.8-1.4      8.4-14.3
Chiron 180          105          46             1.4          15.8
                                34-65         1.1-1.8      10.4-24.3
DPC Immulite         41          7.7            0.3           2.8
                               6.8-8.9        0.1-0.5       2.4-3.4
Hybritech           194         18.9            0.9           6.8
Tandem-E                        16-21         0.7-1.0       6.0-7.7
Tosoh AIA Pack      116         13.2            0.6           4.9
                              11.0-15.5       0.5-0.8       4.2-5.5

n *: indicates the number of tests done on sample 1

Data survey K/KNA Ligand Survey for CAP 1997.

At this time there are no data to explain the fact that some PSA is not complexed (due to an excess of ACT) or that patients with prostate cancer generally have higher levels of this complex form.

Studies comparing fPSA to PSA indicate that in untreated patients, the fPSA assay detects prostate cancer better than PSA.[32-37] In a recent multisite study of 773 men with PSA results between 4 and 10 ng/mL, PSA was compared with fPSA. Measurement of fPSA reduced unnecessary biopsies and the number of false positives (specificity for PSA 46% vs. 78% for fPSA).[32] While much of the data for fPSA are encouraging, the cut-off values in the literature range from 14% to 28%.[33-37] This is due to several factors, including previously mentioned variations between assays. The general opinion seems to be that percent fPSA adds useful information for identifying men with prostate cancer.[18]

Table 3

Age-specific PSA ranges(*)

Age                            PSA range
(yrs)                            (ng/mL)

40-49                                0.0-2.5
50-59                                0.0-3.5
60-69                                0.0-4.5
70-79                                0.0-6.5

* Values are based on the Hybritech Tandem E method.

Ultrasensitive assays for PSA. Since 1993, ultrasensitive detection of lower limits of PSA (0.002-0.01 ng/mL) has been available for research and commercial methods.[35-40] Ultrasensitive methods could have important clinical applications for early detection of postsurgical relapse or residual prostate cancer.[40] By detecting values in a lower range, ultrasensitive methods might help to determine PSA velocity more accurately and consequently, detect prostate cancer earlier.[39] They might also be useful for monitoring prostatectomized patients to ensure that prostate cancer is not recurring and secreting low levels of PSA.[41] Diamadis agreed that the ultrasensitive assay could detect a relapse 1 year earlier than assays with limits of 0.1 ng/mL,[42] but Junker et al. were not able to confirm these data.[44]

Multiple markers and neural networks. In 1996 Stamey applied neural networks as a tool for evaluating men at risk for prostate cancer.[43] The laboratory data for this approach included age and values for PSA, CK isoenzymes, and PAP. From these data, the neural network calculates an index called ProstAsure. In one multisite study of this index, its sensitivity and specificity were 81% and 93%, respectively, compared to 80% and 74% for fPSA. If the sensitivity for fPSA were increased to 93%, the specificity dropped to 59%. Thus, there were fewer false-positive values for the Prostasure index than with fPSA.45 When this index was evaluated in patients with a PSA less than 4.0 ng/mL, it had a higher specificity (57%) than fPSA (38%) at a sensitivity of 90%.[46] Other data[46-48] suggest that ProstAsure may be valuable for patients whose PSA is less that 4.0 ng/mL.

Future assays, Research continues to improve the ability of the laboratory to aid the clinician in detecting prostate cancer. Recent research[49-54] with polymerase chain reaction (PCR) and reverse transcriptase-PCR (RT-PCR) has suggested that micrometastatic prostate cancer cells can be detected by these methods. These early results indicate that this approach could be a promising aid to the diagnosis and management of prostate cancer.[55]

Monitoring. After treatment for prostate cancer, PSA levels should decline. Since the half-life of PSA is approximately 2 to 3 days, PSA values near zero should be detected shortly after radical prostatectomy if the disease is confined to the prostate alone. If radiation is used instead of radical prostatectomy, PSA values also decline, but values near zero are less common, probably because the remaining normal tissue still secretes PSA. In 40% to 75% of patients treated surgically or with radiation, PSA values stabilize in the normal range within 6 to 12 months.[4,8]

Conclusions

Serum PSA levels have been a valuable aid for early diagnosis of prostate cancer. However, used alone, PSA values are neither specific nor sensitive enough to use for early diagnosis and staging of prostate cancer.

The use of PSA density, PSA velocity, and age-related normal values may improve the value of total PSA results. Age-specific normal ranges appear to be more sensitive for men between the ages of 50 and 60 years. However, use of PSA density and PSA velocity has limitations that include the following:

* To use PSA density, prostate volume must be accurately determined by TRUS, but the accuracy of this procedure is examiner-dependent and more costly than PSA or DRE.

* PSA velocity seems to be more sensitive and specific than PSA alone. However, PSA velocity requires at least three PSA measurements over a two-year period. Most clinicians recommend an annual assay for PSA in men over the age of 50.[2-4] Therefore, parameters needed for PSA velocity can be obtained. However, should any of the values appear abnormal (greater than 4 ng/mL), further (and perhaps unnecessary) testing would be indicated.

* Percent fPSA appears to be useful for detecting prostate cancer when the total PSA is between 4 and 10 ng/mL.

* Using the ProstAsure index may add diagnostic value when PSA levels are between 4 and 10 ng/mL.

* Newer tools, such as RT-PCR, may enable the early detection of PSA mRNA. However, results are contradictory, and more data from this technique are needed before a consensus can be reached.

All these conclusions are valid, but none address the question of whether prostate cancer should be treated after it is identified (see Sidebar). Currently, there is little agreement on this, but it would appear that the patient should decide whether to proceed with treatment after the physician provides information on:

* the effect of the tumor on the patient's life expectancy, compared to the patient's life expectancy if there were no tumor.

* the treatments available for the tumor, as well as the life expectancy following treatment.

* the side-effects of the treatments.

* the costs of the treatment options and the time to recover from the treatment.

These are exciting times for PSA assays and the detection of prostate cancer. Current approaches for evaluating prostate disease are not perfect. However, research continues to improve the utility of the various assays for PSA and to evaluate other approaches for diagnosis and management of prostate cancer. This is good news as life expectancy increases and the population ages.

Table 1

How PSA levels change

Factors that increase PSA levels

Prostate disease

* cancer

* prostatic interstitial neoplasia

* benign prostatic hyperplasia

* prostatitis

* prostate ischemia

Clinical manipulations

* digital rectal exam

* transurethral ultrasound

* prostate biopsy

* indwelling urinary catheter

Treatments

* prostate surgery

* radiation therapy

Extra-prostate sources

* parotid glands

* periurethral glands

Other diseases

* acute renal failure

* acute myocardial infarction

* bypass surgery

Analytical interferences

* antibody cross-reactivity

* high-titer heterophile antibodies

* increased levels of kallikrein

Other

* patient age and size of gland

* testosterone administration

* LH/FSH administration (transient effect)

Factors that decrease PSA levels

* antiandrogen drugs

* hospitalization

* castration

* ejaculation

* prostatectomy

* radiation therapy

* analytical factors including artificially low results due to extremely high levels of PSA (hook effect) or improper specimen collection

The author wishes to thank Lety Hernandez and Spencer Schaefer, without whose help this article would not have been possible.

References

1. Wingo P, Landis S, Ries L. An adjustment to the 1997 estimate for new prostate cancer cases. CA Cancer J Clin. 1997;47:239-242.

2. Cerosimo R, Chan D. Prostate cancer: Current and evolving strategies. Am J Health Sys Pharm. 1996;53:381-396.

3. Giovannucci E. How is individual risk for prostate cancer assessed? Hematology/Oncology Clinics N Am. 1996;10:537-548.

4. Buckman R. What you really need to know about cancer. Baltimore, Maryland: Johns Hopkins University Press; 1997.

5. Armbruster D. Prostate specific antigen: Biochemistry, analytical methods and clinical application. Clin Chem. 1993;39:181-195.

6. Wang M, Valenzuela L, Murphy G, et al. Purification of a human prostate specific antigen. Invest Urol. 1979; 17:159-163.

7. Abrahamsson PA, Lilja H, Oesterling J. Molecular forms of prostate specific antigen. Urology Clin N Am. 1997;24:353-365.

8. Pannek J, Partin A. Prostatic specific antigen: What's new in 1997. Oncology. 1997;11:1273-1280.

9. Van Duijnhoven H, Pequeriaux N, Van Zon J, et al. Large discrepancy between prostatic specific antigen from different assays during longitudinal follow-up. Clin Chem. 1996;42:637-641.

10. Wener M, Daum P, Brawer M. Variation in measurement of prostatic specific antigen: Importance of method and lot variability. Clin Chem. 1995;41:1730-1773.

11. Semjonow A, Oberpenning F, Brandt B, et al. Impact of free prostatic specific antigen on discordant measurement results of assays for total prostatic specific antigen. Urol. 1996;48:10-15.

12. Kadmon D, Weinberg A, Williams R, et al. Pitfalls in interpreting prostatic specific antigen velocity. J Urol. 1996;155:1655-1657.

13. Balducci L, Pow-Sang J, Friedland J, Diaz J. Prostate cancer. Clinics in Geriatric Medicine. 1997;13:283-306.

14. Catalona W, Richie J, Ahmann, et al. Comparison of Digital Rectal Examination and Serum Prostate Specific Antigen. J Urol. 1994;151:1283-1290.

15. Catalona W. Prostate cancer. In: Reintgen D, Clark R, eds. Cancer Screening. St. Louis, MO: Mosby; 1996.

16. Carter H, Pearson J, Metter J, et al. Longitudinal evaluation of prostate specific antigen levels in men with and without prostate disease. JAMA. 1992;267:2215-2220.

17. Speights V, Brawn P, Foster D, et al. Evaluation of age specific normal ranges for prostatic specific antigen. Urol. 1995;46:200-204.

18. Partin A, Criley S, Subong E, et al. Standard versus age-specific prostatic specific antigen in screening prostate adenocarcinoma. J Urol. 1996; 155:1336-1339.

19. Oesterling J, Jacobson S, Chute C, et al. Prostatic specific antigen in a community-based population of healthy men. JAMA. 1993;270:860-864.

20. Oesterling J, Cooner W, Jacobson S. et al. Influence of patient age on the serum PSA concentration. Urol Clin N Am. 1993;20: 671-680.

21. Catalona W, Hudson M, Scardino P, et al. Selection of optimal PSA cutoff for early detection of prostate cancer. J Urol. 1994; 152:2037-2042.

22. Zlotta A, Djavan B, Marberger M, et al. Prostatic specific antigen density of the transition zone. J Urol. 1997;157:1315-1321.

23. Beduschi M, Oesterling J. Prostatic specific antigen density. Urol Clin N Am. 1997;24:323-332.

24. Bensen M, Whang I, Pantuck A, et al. Prostatic specific antigen density. J Urol. 1992;147:815-816.

25. Henderson R, Easham J, Culkin D, et al. Prostatic specific antigen and prostatic specific antigen density. J Natl Cancer Inst. 1997; 89:134-138.

26. Seaman E, Whang I, Olsson C, et al. PSA density: Role in patient evaluation. Urol Clin N Am. 1993;20:653-663.

27. Cookson M, Floyd M, Ball T, et al. Lack of Predictive value of prostatic specific antigen density in the detection of prostate cancer. J Urol. 1995; 154:1070-1073.

28. Beduschi M, Oesterling J. Prostatic specific antigen density. Urol Clin N Am. 1997;24:323-332.

29. Carter H, Pearson J. Prostatic specific antigen velocity. Urol Clin N Am. 1997;24:333-338.

30. Oesterling J, Suman V, Sincke H, et al. PSA-detected prostate cancer. Urol Clin N Am. 1993;20:687.

31. Smith D, Catalona W. Rate of change in prostatic specific antigen level as a method for prostate cancer detection. J Urol. 1995; 152:1163-1167.

32. Catalona W, Partin A, Slawin K, et al. Use of the percentage of free prostatic specific antigen to enhance differentiation of prostate cancer from benign prostate disease. JAMA. 1998;279:1542-1547.

33. Catalona W, Smith D, Wolfert R, et al. Evaluation of percentage of free prostatic specific antigen to improve specificity of prostate cancer screening. JAMA. 1995;274:1214-1220.

34. Morgan T, McLeod D, Leifer E, et al. Prospective use of free prostatic specific antigen to avoid repeat prostate biopsies. Urol. 1996;48:76-80.

35. Egawa S, Soh S, Ohori M, et al. The ratio of free to total serum prostatic specific antigen and its use in differential diagnosis of prostate carcinoma in Japan. Cancer. 1997;79:90-98.

36. Luderer A, Chen Y-T, Soriano T, et al. Measurement of the proportion of free to total prostatic specific antigen improves diagnostic performance. Urol. 1995;46:187-194.

37. Oesterling J, Jacobson S, Klee G, et al. Free, complexed and total serum prostatic specific antigen: The establishment of appropriate reference ranges. J Urol. 1995; 154:1090-1095.

38. Stamey T, Graves H, Wehner N, et al. Early detection of residual prostate cancer after radical prostatectomy by an ultrasensitive assay for prostatic specific antigen. J Urol. 1993;149:516-518.

39. Groves H, Wehner N, Stamey T. Ultrasensitive RIA for prostatic specific antigen. Clin Chem. 1992;38:735-742.

40. Yu H, Diamandis E. Ultrasensitive time-resolved immunofluorometric assay of prostatic specific antigen. Clin Chem. 1993;39:2108-2114.

41. Mione R, Marichello M, Sartorello P, et al. Third-generation prostatic specific antigen. Int J Biol Markers. 1995;1:229-235.

42. Diamandis E. Clinical application of ultrasensitive prostatic specific antigen Assays. J Nat Cancer Inst. 1997;89:1077.

43. Stamey T, Barnhill S, Zhang Z, et al. Effectiveness of ProstAsure in detecting PCA and BPH in men age 50 and older. J Urol. 155: Suppl Abstract 504, May 1996.

44. Junker R, Brandt B, Semjonow A, et al. Measurement of prostatic specific antigen: No advantage to ultrasensitive assays? J Nat Cancer Inst. 1997;88:1594.

45. Stamey T, Barnhill S, Zhang Z, et al. Comparison of a neural network with high sensitivity and specificity to free/total serum prostatic specific antigen for diagnosing prostate cancer in men with prostatic specific antigen greater than 4.0 ng/mL. In: Stamey, T, ed. Monographs in Urology. Monteverde, FL: Medical Directions Publishing, Inc. 1998;19:23-32.

46. Barnhill S, Stamey T, Zhang Z, et al. The ability of the ProstAsure index to identify patients with low cancer volumes and high potential for cure [abstract]. Am Urol Assoc. 1997; Abstract 241.

47. Stamey T, Barnhill S, Zhang Z, et al. A neural network with high sensitivity and specificity for diagnosing cancer in men with prostatic specific antigen greater than 4.0 ng/mL [abstract]. Am Urol Assoc. 1997; Abstract: 1425.

48. Barnhill S, Zhang Z, Madyastha K, et al. The Prostasure index is useful in the diagnosis of early prostate cancer but does not predict pathological stage in men with as serum prostatic specific antigen less than 4.0 ng/mL [abstract]. Am Urol Assoc. 1997; Abstract: 1809.

49. Katz A, Olsson C, Raffo A, et al. Molecular staging of prostate cancer with the use of an enhanced RT-PCR assay. J Urol. 1994;43;765-775.

50. Ignatoff J, Oelfelein M, Herz B, et al. Prostatic specific antigen RT-PCR assay in preoperative staging of PCA. J Urol. 1996;155:417A.

51. Sokoloff M, Tso C, Kaboo S, et al. Quantitative PCR does not improve perioperative PC A staging. J Urol. 1996, 156:1560-1566.

52. Slawin K, O Hara S, Song W, et al. Comparison of results using different RT-PCR assay methods on a single set of specimens. J Urol. 1996;155:417A.

53. Edelstein I, Zeitman A, de las Morenas A, et al. Implications of prostatic micrometasteses to the pelvic lymph node. J Urol. 1996;155:550A.

54. Ellis W, Vassella R, Corey E, et al. Use of a RT-PCR to detect prostate cancer cells in the blood. J Urol. 1996;155:378A.

55. Nejat R, Katz A, Olsson C. The role of RT-PCR for staging patients with clinically localized PCA. Seminars in Urologic Oncology. 1998;16:40-45.

56. Wallace DM, Chisholm GD, Hendry WF. TNM classification for urological tumors (UICC) 1974. Br J Urol. 1975;47:1-12.

57. Gleason DS, Mellinger GT. Veterans Administration Cooperative Urological Research Group: Prediction of prognosis for prostatic carcinoma by combined histological grading and staging. J Urol. 1974; 111:58:67.

RELATED ARTICLE: Staging, grading, and treatment of prostate cancer

Once prostate cancer is definitively diagnosed, additional tests, including biopsies, will be used to determine the stage (severity) of the disease. Two systems for staging prostate cancer include the American Urological System (AUS), which designates four major stages A through D, and the American Joint Committee for Cancer Staging and End Result Reporting[56] which uses the tumor, lymph nodes, and metastases (TNM) approach.

For example, a tumor confined within the prostate capsule with no regional lymph node metastases and no evidence of distant metastases would be staged B with the AUS system and [T.sub.2][N.sub.0][M.sub.0] using the TNM system. In stage C or [T.sub.3][N.sub.0][M.sub.0], the cancer would be a palpable tumor extending beyond the patient's prostate gland but without metastases to regional lymph nodes or distant sites.

Histologic grading of the tumor is also part of the initial examination of the newly diagnosed patient. In a scheme developed by Gleason,[57] tumor cell growth patterns and degrees of differentiation are studied and scored. The higher the overall grade of a tumor, the more undifferentiated or juvenile cells it contains. For example, a slow growing tumor containing well-differentiated cells would have a score of 2 to 4, while a tumor with poorly differentiated cells would receive a score of 8 to 10. The Gleason score gives the best correlation with the mortality rate in a localized disease. The higher the score, the worse the prognosis.[13]

In a study of over 20,000 patients with new#y diagnosed disease, 29% were diagnosed in stage A (lowest grade, most successfully treated), 38% in stage B, 12% in stage C, and 21% in stage D.[15] Between 1974 and 1990, prostate cancer was diagnosed earlier; stages A and B were diagnosed in 57% of patients in 1974 versus 67% in 1990.

Treatment

The most common treatment for prostate cancer is radical prostatectomy, but recently, external beam irradiation or prostate gland implantation with radioactive "seeds" of iodine or palladium have been used. Sometimes both external and internal irradiation are employed, with external beam irradiation preceding seed therapy. Castration is sometimes performed and can be achieved either surgically or medically (with hormones). The aim of such treatment is to prevent androgenic stimulation of prostate cancer growth. Surgical castration is less costly and rapidly lowers the level of circulating testosterone. Treatment with hormones, such as the estrogen diethylstilbestrol (DES), or with an LH-RH (luteinizing hormone-releasing hormone) analog such as leuprolide acetate (Lupron) can suppress testosterone production to castration levels. This approach is more costly and the decline in testosterone level is more gradual than in the surgical approach. The mechanism by which androgens stimulate the growth of prostate cancer is not clear.[4,13]

Treatment of stage [T.sub.1] (A). At this stage, the tumor is neither visible by TRUS nor palpable by DRE. If the tumor is detected at this stage, it is detected either by PSA screening or as a result of another procedure.[15] Treatment options include observation, surgery, or radiation.

Treatment of stage [T.sub.2] (B). At this stage, prostate cancer is confined to the gland and clinically detectable using DRE, PSA, and TRUS. These patients are usually candidates for radical prostatectomy or radiation therapy (either external beam or seed implants). The 5-year survival rates for these three treatments are comparable (greater than 70%), considering that (1) the stage assigned is more definite when surgery is performed, (2) a thorough pathological exam of the tumor tissue is conducted at the time of the surgery, and (3) clinicians have preferences based on their background and training (most urologists recommend surgery; most radiologists recommend radiation).[4]

Treatment of stage [T.sub.3] or [T.sub.4] (C). This cancer is locally advanced beyond the capsule. There are no data to support surgery for these patients, and the primary treatment is external beam radiation. The 5-year survival rates are approximately 55%.[4]

Treatment of stage N (spread to lymph nodes) and M (spread to distant organs) (D). This stage is metastatic beyond the gland, usually to lymph nodes and bone. Two common treatment approaches for these patients include only observation or administration of hormones. For those patients with neither symptoms nor local lesions that require palliative treatment, observation may be the choice. Delaying treatment of these patients does not affect their ultimate survival time and can spare them the expense and side effects of hormone therapy.

David Plaut is a scientific specialist at Dade Behring in Newark, DE.

COPYRIGHT 1998 Nelson Publishing
COPYRIGHT 2004 Gale Group