Primary Prevention
Screening
        Prostate-specific antigen and digital rectal exam
        Candidate prostate cancer biomarkers
        Multiplex assays
Treatment

Refer to the PDQ summaries on Screening for Prostate Cancer; Prevention of Prostate Cancer; and Prostate Cancer Treatment for more information on interventions for sporadic nonfamilial forms of prostate cancer.

As with any disease process, decisions about risk-reducing interventions for patients with an inherited predisposition to prostate cancer are best guided by randomized controlled clinical trials, and by knowledge of the underlying natural history of the process. Unfortunately, little is known about either the natural history or the inherent biologic aggressiveness of familial prostate cancer compared with sporadic forms. Existing studies of the natural history of prostate cancer in men with a positive family history are predominantly based on retrospective case series. Because awareness of a positive family history can lead to more frequent work-ups for cancer and result in apparently earlier prostate cancer detection, assessments of disease progression rates and survival after diagnosis are subject to selection, lead time, and length biases. Refer to the PDQ summary on Cancer Screening Overview for more information.

Given the paucity of information on the natural history of prostate cancer in men with a hereditary predisposition, decisions about risk reduction, early detection, and therapy are currently based on the literature used to guide interventions in sporadic prostate cancer, coupled with the best clinical judgment of those responsible for the care of these patients, with the active participation of well-informed high-risk patients.

There are no definitive studies of primary prevention strategies in men with a hereditary risk of prostate cancer. Thus, there are no definitive recommendations that can be offered to these patients to reduce their risk for prostate cancer at the present time.

The Prostate Cancer Prevention Trial (PCPT), a prospective, randomized clinical trial of finasteride versus placebo, demonstrated a 25% reduction in prostate cancer risk among study participants receiving finasteride.[1] Finasteride administration produced statistically similar reductions in prostate cancer risk in family history positive (19% decrease) and family history negative (26% decrease) subjects. A subsequent PCPT publication suggested that end-of-study biopsies in asymptomatic men with serum prostate-specific antigen (PSA) values consistently lower than 4.0 ng/mL were more likely to detect prostate cancer in men with an affected first-degree relative (19.7%) versus those with a negative family history (14.4%).[1]

The concern over the reported increase of high-grade prostate cancer in the finasteride arm compared with the placebo arm (6.6% of men analyzed vs. 5.1%, respectively, P = 0.005) in the original report from the PCPT was recently reanalyzed with consideration of possible biases that may have influenced these findings.[2] These biases included improved sensitivity of the PSA and digital rectal exam (DRE) for overall prostate cancer detection with finasteride, improved sensitivity of PSA for high-grade prostate cancer detection with finasteride, differences in participants reaching the study endpoints between the two arms, and increased detection of high-grade disease with finasteride due to reduction in size of the prostate gland. Using a bias-adjusted modeling analysis, 7,966 participants in the finasteride arm and 8,024 participants in the placebo arm of the PCPT were studied. No statistically significant difference was found in the overall prevalence of high-grade prostate cancer with finasteride compared with placebo (4.8% vs. 4.2%, respectively, P = 0.12). Further analysis in a subset of men with a prostate cancer diagnosis who were treated with radical prostatectomy (n = 500) revealed that men on finasteride had less high-grade prostate cancer compared with men who took placebo (6.0% vs. 8.2%, respectively). The estimated risk reduction for high-grade prostate cancer from this subset analysis in men who had a prostatectomy and took finasteride was 27% (relative risk [RR], 0.73; 95% confidence interval [CI], 0.56–0.96; P = 0.02).[2]

Another study estimated the rate of true high-grade prostate cancer in the PCPT by extrapolating the Gleason score from the subset of participants who had a radical prostatectomy.[3] Statistical modeling that accounted for misclassification of Gleason score from biopsy to radical prostatectomy was used in this study. When comparing finasteride with placebo, the estimated RR for low-grade and high-grade prostate cancer at prostatectomy was 0.61 (95% CI, 0.51–0.71) and 0.84 (95% CI, 0.68–1.05), respectively. Information was not reported as to whether men with a family history of prostate cancer had a reduction in high-grade prostate cancer in these analyses. Further definition of the prostate cancer prevention potential of finasteride in men with a family history of prostate cancer, along with genetic stratification to identify those men at truly increased risk for the disease, remains to be determined. Together these two studies suggest that the apparent excess risk of high-grade prostate cancer in men treated with finasteride is not substantiated. Rather, the apparent association may be explained by various biases not accounted for in the original analysis.

Level of Evidence: Iaii

Refer to the PDQ summary on Prevention of Prostate Cancer for a more detailed description of the prevention of prostate cancer in the general population. Information about ongoing prostate cancer prevention clinical trials is available from the NCI Web site.

There is little information about the net benefits and harms of screening men at higher risk of prostate cancer. There is no evidence to support specific screening approaches in prostate cancer families at high risk. Risks and benefits of routine screening in the general population are discussed in the PDQ summary on Screening for Prostate Cancer.

There is limited information about the efficacy of commonly available screening tests such as the DRE or serum PSA in men genetically predisposed to developing prostate cancer. Furthermore, comparing the results of studies examining the efficacy of screening for prostate cancer is difficult; studies vary with regard to the cut-off values chosen for an elevated PSA test. For a given sensitivity and specificity of a screening test, the positive predictive value (PPV [proportion of men with positive tests who have prostate cancer]) increases as the underlying prevalence of disease rises. Therefore, it is theoretically possible that the PPV and diagnostic yield will be higher for the DRE and for PSA in men with a genetic predisposition than in average-risk populations.[4,5]

Currently, there are only a few case-control studies and no published randomized trials examining screening in men with an increased risk of prostate cancer. A 10-year longitudinal study of serum PSA and DRE every 6 to 12 months in high-risk men older than 40 years has been conducted.[6] Two high-risk categories (1,227 men with a family history of prostate cancer and 1,224 African American men) were compared with 15,964 low-risk non–African American men without a family history of prostate cancer. Suspicious screening results were present in 7% of non–African American men with a family history of prostate cancer, 8% of the low-risk African American men, and 20% of African American men with a family history of prostate cancer. The PPV was inversely proportional to age for those who had an abnormal screening test and underwent biopsy. Among men aged 40 to 49 years, the PPV was 50% for non–African American men with a positive family history, 54% for African American men without a family history, and 75% among African American men with a family history, compared with 38%, 49%, and 52%, respectively, among men aged 50 years and older. Of the 16 cancers detected in high-risk men younger than 50 years, 15 were clinically significant, with intermediate Gleason scores (5–7), and three were not confined to the prostate.[6]

One screening study of the relatives of 435 men with prostate cancer measured serum PSA every 12 months for 2 years. Four-hundred and forty-two participants were classified into two groups: sporadic (defined as only one first-degree relative with prostate cancer) or familial (with two or more cases of prostate cancer). PSA higher than 0.004 mg/L was present in 0.8% in men aged 40 to 49 years, compared with 12.4% of men older than 50 years. No differences in prostate cancer detection rates or elevated PSA levels were found between sporadic and familial groups. Of the ten prostate cancers detected in this study, nine were clinically localized and of intermediate Gleason scores (5–7).[7]

In a Finnish prostate cancer screening study, family history of prostate cancer was obtained in 2,099 prostate cancer patients.[5] This resulted in the identification of 103 prostate cancer families with two or more affected first-degree or second-degree relatives having at least one living first-degree unaffected male. From those families, 209 of 226 eligible first-degree unaffected asymptomatic males aged 45 to 75 years were enrolled in a study involving a single serum PSA measurement. An elevated PSA (2.6–28.3 mg/L) was identified in 21 (10%) of subjects. Subsequent biopsies revealed prostate adenocarcinoma in seven (3.3%), including one at an advanced stage, and two in prostatic intraepithelial neoplasia (1%). The mean age of PSA-detected cancers was 65.1 years, 7 years younger than the average age of prostate cancer diagnosis in Finland. In men with a family history of early-onset prostate cancer (mean age of diagnosis in the family <60 years), the frequency of elevated PSAs was 28.6% and subclinical prostate cancer was 14.3%, significantly higher than the 2.3% to 4.5% reported in other PSA screening studies of this type.[8-13] These findings, however, may not be comparable to U.S. studies: prostate screening practices may differ between Finland and the United States, and rates of prior screening in the population studied were not reported.

A large French Canadian study reported findings from 6,390 men older than 45 years who underwent prostate screening consisting of annual serum PSA and DRE followed by transrectal ultrasound imaging if an abnormality was detected. Of these, 1,563 (24.5%) were found to have an abnormal rectal exam (n = 504) or a PSA above 3.0 mg/L (n = 1,261).[13] Twenty-six refused follow-up; of the remaining subjects, 50.5% underwent biopsy following ultrasound examination. Prostate cancer was identified in 264 men, representing 34.0% of those who underwent biopsy and 4.1% of all 6,390 enrolled subjects. The prevalence of screen-detected prostate cancer was highest in men reporting a brother with prostate cancer (10.21%), as opposed to those reporting a father with prostate cancer (4.75%). Overall in this study, the PPV of a PSA more than 3.0 mg/L was significantly associated with a family history. The PPV was 28.6% in men with a prostate cancer family history and 17.9% in men without an affected first-degree relative. The increase in PPV of PSA was confined to the men with a normal rectal exam.[13]

A PSA screening study of 20,716 asymptomatic men identified by the Finnish population-based registry did not find a higher PPV for men with a family history of one or more first-degree relatives with prostate cancer, compared with controls. Using a PSA cut-off of 0.004 mg/L, the PPV of an abnormal PSA for the 964 men with a positive family history was 28% vs. 31% for the 19,347 men without a family history. The RR of developing prostate cancer among male relatives of men with prostate cancer was modest (RR = 1.3; 95% CI, 0.95–1.71), suggesting that the family history was not a significant prostate cancer risk factor in this study. This unexpected finding might account for the lack of differences seen in the PPV of the PSA test when comparing men with and without a family history of prostate cancer.[14]

Current recommendations for screening at-risk members of familial or hereditary prostate cancer kindreds are based on expert opinion panels.[15,16] Therefore, the overall summary of evidence related to the efficacy of screening is level 5. There are no randomized studies that address screening at-risk members of familial or hereditary prostate cancer kindreds, and the observational data are contradictory. Refer to the Screening Behaviors section of this summary for more information on factors that influence prostate cancer screening.

Level of Evidence: 5

Many new prostate cancer biomarkers (either alone or in combination) will be identified and proposed during the next 5 to 10 years. While this is an active area of research with a number of promising new biomarkers in early development, none of these biomarkers alone or in combination have been sufficiently well-studied to justify their routine clinical use for screening purposes either in the general population or in men at increased risk of prostate cancer based on family history.

Before addressing information related to emerging prostate cancer biomarkers, it is important to consider the several steps that are required to develop and, more importantly, to validate a new biomarker. One useful framework was described by the National Cancer Institute (NCI) Early Detection Research Network investigators.[17] These authors indicated that the goal of a cancer-screening program is to detect tumors at an early stage so that treatment is likely to be successful. The gold standard by which such programs are judged is whether the death rate from the cancer for which screening is performed is reduced among those being tested. In addition, the screening test must be sufficiently noninvasive and inexpensive to allow widespread use. Maintaining high test specificity (i.e., few false-positive results) is essential for a population screening test, because even a low false-positive rate results in many people having to undergo unnecessary and costly diagnostic procedures and psychological stress. It is likely that the use of several such cancer biomarkers in combination will be required for a screening test to be both sensitive and specific. Furthermore, a clinically useful test must have a high PPV (a parameter derived from sensitivity, specificity, and disease prevalence in the screened population). Practically speaking, a biomarker with a PPV of 10% implies that ten surgical procedures would be required to identify one case of prostate cancer; the remaining nine surgeries would represent false-positive test findings.[18] In general, the prostate cancer research community considers biomarkers with a PPV less than 10% to be clinically unacceptable. Finally it is important to keep in mind that while novel biomarkers may be present in the sera of men with advanced prostate cancer (which comprise the vast majority of cases analyzed in the early phases of biomarker development), they may or may not be detectable in men with early stage disease. This is essential if the screening test is to be clinically useful in the detection of localized and potentially curable prostate cancer.

It has been suggested that there are five general phases in biomarker development and validation:[17]

Phase I — Preclinical exploratory studies

• Identify potentially discriminating biomarkers.
• Usually done by comparing gene over- or under-expression in tumor compared with normal tissue.
• Since many exploratory analyses in large numbers of genes are performed at this stage, one or more may seem to have good discriminating ability between cancers and normal tissue by random chance alone.

Phase 2 — Clinical assay development for clinical disease

• Develop a clinical assay that can be obtained on noninvasively obtained samples (e.g., a blood specimen).
• Often the test targets the protein product of one of the genes found to be of interest in phase I.
• The goal is to describe the performance characteristics of the assay for distinguishing between subjects with and without cancer. At this point, the assay should be in its final configuration and remain stable throughout the following phases.
• IMPORTANT: Since the case subjects in a phase 2 study already have cancer, with assay results obtained at the time of disease diagnosis, one cannot determine if disease can be detected early with a given biomarker.

Phase 3 — Retrospective longitudinal repository studies

• Compare clinical specimens collected from cancer case subjects before their clinical diagnosis with specimens from subjects who have not developed cancer.
• Evaluate, as a function of time before clinical diagnosis, the biomarker’s ability to detect preclinical disease.
• Define the criteria for a positive screening test in preparation for phase 4.
• Explore the influence of other patient characteristics (e.g., age, gender, smoking status, medication use) on the ability of the biomarker to discriminate between those with and without preclinical disease.

Phase 4 — Prospective screening studies

• Determine the operating characteristics of the biomarker-based screening test in a population for which the test is intended.
• Measure the detection rate (number of abnormal tests among all those with the disease) and the false-positive rate (the number of abnormal tests among all those who do not have the disease).
• Evaluate whether the cancers detected by the test are being found at an early stage, a point at which treatment is more likely to be curative.
• Assess whether the test is acceptable in a population of persons for whom it is intended. Will subjects comply with the test schedule and results?

Phase 5 — Cancer control studies

• Ideally, randomized controlled clinical trials in clinically relevant populations, in which one arm is subjected to screening and appropriate intervention if screen-positive, while the other arm is not screened.
• Determine whether the death rate of the cancer being screened for is reduced among those who use the screening test.
• Obtain information about the costs of screening and treatment of screen-detected cancers.

Finally, for a validated biomarker test to be considered appropriate for use in a particular population, it must have been evaluated in that specific population without prior selection of known positives and negatives. In addition, the test must demonstrate clinical utility, that is, a positive net balance of benefits and risks associated with the application of the test. These may include improved health outcomes, as well as net psychosocial and economic benefits.[19]

Prostate cancer poses a further challenge relative to the potential impact of false-positive test results. There are no reliable noninvasive diagnostic tests for early-stage disease, and the value of identifying early-onset disease has not been established. This is further complicated by the fact that prostate cancer is clinically heterogeneous, i.e., a proportion of prostate cancer may be relatively indolent disease of uncertain clinical significance.[18] High test specificity (i.e., a very low false-positive rate) is required to avoid unnecessary screening and further diagnostic evaluation, which may include surgery.

New candidate prostate cancer SNPs have been identified and studied individually, in combination with family history, or in various other permutations. Most of the study populations are relatively small and comprise highly-selected known prostate cancer cases and healthy controls of the type evaluated in early development phases I and II. Results have not been consistently replicated in multiple studies; presently, none are considered ready for widespread clinical application.

Because individual single nucleotide polymorphisms (SNPs) have not met the criteria for an effective risk assessment test, it has been suggested that testing multiple prostate cancer-related SNPs may be required to obtain satisfactory results. An initial study evaluated five chromosomal regions associated with prostate cancer in a Swedish population, three at 8q24, one at 17q12 and one at 17q24.3.[20] Sixteen SNPs within these regions were assessed in 2,893 men with prostate cancer and 1,781 controls. It was estimated that the five SNPs most strongly associated with prostate cancer accounted for 46% of prostate cancer in the Swedish men from this study. When considered independently, each SNP was associated with a small increase in prostate cancer risk. However, the investigators identified a cumulative stronger association with prostate cancer risk when multiple SNPs and family history were combined, versus men without any risk SNPs or a prostate cancer family history.[20]

A larger study of 5,628 men with prostate cancer and 3,514 controls from the United States and Sweden, further strengthened this association.[21] For men carrying one or more risk SNPs, the estimated OR (95% CI) was 1.41 (1.20–1.67) for one SNP to as high as 3.80 (2.77–5.22) for four or more SNPs. The cumulative effect of family history with up to five SNPs was estimated to have an odds ratio (OR) (95% CI) of 11.26 (4.74–24.75) for prostate cancer.[21] The observation that family history added significant strength to the SNP-related association suggests that there may be additional genetic risk variants yet to be discovered. All available data to date are derived from studies of sporadic prostate cancer. Familial prostate cancer has not been evaluated.

Viewed in the context of the criteria previously described, this five-SNP assay would be classified as phase 2 in its development. While this appears to be a promising avenue of prostate cancer risk evaluation, additional validation is required, particularly in an unselected population representative of the clinical population of interest.

Various studies have shown better, worse, or similar survival rates after treatment in men with prostate cancer who have a family history of affected first-degree relatives, compared with those who have a negative family history.[22-25] There is extensive literature addressing whether family history of prostate cancer is linked with aggressive tumor behavior and consequently a worse prognosis. The most current longitudinal report suggests that this is not likely the case.[26]

In general, there is insufficient information available to determine whether treatment strategies differ in efficacy for sporadic cases versus familial cases of prostate cancer. Decisions about treating familial cases of cancer are currently guided by information derived from therapeutic studies in the general population of prostate cancer patients. Therefore, no level of evidence is assigned. A detailed discussion of treatment in these patients is found in the PDQ Prostate Cancer treatment summary, and information about ongoing prostate cancer treatment clinical trials is available from the NCI Web site.

Level of Evidence: Not assigned

References

1. Thompson IM, Goodman PJ, Tangen CM, et al.: The influence of finasteride on the development of prostate cancer. N Engl J Med 349 (3): 215-24, 2003.  [PUBMED Abstract]
   
   
2. Redman M, Tangen C, Goodman P, et al.: Finasteride does not increase the risk of high-grade prostate cancer: a bias-adjusted modeling approach. Cancer Prev Res Phila Pa 1 (3): 174-81, 2008. 
   
   
3. Pinsky P, Parnes H, Ford L: Estimating rates of true high-grade disease in the prostate cancer prevention trial . Cancer Prev Res Phila Pa 1 (3): 182-6, 2008. 
   
   
4. Sartor O: Early detection of prostate cancer in African-American men with an increased familial risk of disease. J La State Med Soc 148 (4): 179-85, 1996.  [PUBMED Abstract]
   
   
5. Matikainen MP, Schleutker J, Mörsky P, et al.: Detection of subclinical cancers by prostate-specific antigen screening in asymptomatic men from high-risk prostate cancer families. Clin Cancer Res 5 (6): 1275-9, 1999.  [PUBMED Abstract]
   
   
6. Catalona WJ, Antenor JA, Roehl KA, et al.: Screening for prostate cancer in high risk populations. J Urol 168 (5): 1980-3; discussion 1983-4, 2002.  [PUBMED Abstract]
   
   
7. Valeri A, Cormier L, Moineau MP, et al.: Targeted screening for prostate cancer in high risk families: early onset is a significant risk factor for disease in first degree relatives. J Urol 168 (2): 483-7, 2002.  [PUBMED Abstract]
   
   
8. Auvinen A, Tammela T, Stenman UH, et al.: Screening for prostate cancer using serum prostate-specific antigen: a randomised, population-based pilot study in Finland. Br J Cancer 74 (4): 568-72, 1996.  [PUBMED Abstract]
   
   
9. Labrie F, Dupont A, Suburu R, et al.: Serum prostate specific antigen as pre-screening test for prostate cancer. J Urol 147 (3 Pt 2): 846-51; discussion 851-2, 1992.  [PUBMED Abstract]
   
   
10. Standaert B, Denis L: The European Randomized Study of Screening for Prostate Cancer: an update. Cancer 80 (9): 1830-4, 1997.  [PUBMED Abstract]
    
    
11. Catalona WJ, Smith DS, Ratliff TL, et al.: Detection of organ-confined prostate cancer is increased through prostate-specific antigen-based screening. JAMA 270 (8): 948-54, 1993.  [PUBMED Abstract]
    
    
12. Mettlin C, Murphy GP, Babaian RJ, et al.: The results of a five-year early prostate cancer detection intervention. Investigators of the American Cancer Society National Prostate Cancer Detection Project. Cancer 77 (1): 150-9, 1996.  [PUBMED Abstract]
    
    
13. Narod SA, Dupont A, Cusan L, et al.: The impact of family history on early detection of prostate cancer. Nat Med 1 (2): 99-101, 1995.  [PUBMED Abstract]
    
    
14. Mäkinen T, Tammela TL, Stenman UH, et al.: Family history and prostate cancer screening with prostate-specific antigen. J Clin Oncol 20 (11): 2658-63, 2002.  [PUBMED Abstract]
    
    
15. Scardino P: Update: NCCN prostate cancer Clinical Practice Guidelines. J Natl Compr Canc Netw 3 (Suppl 1): S29-33, 2005.  [PUBMED Abstract]
    
    
16. National Comprehensive Cancer Network.: NCCN Clinical Practice Guidelines in Oncology: Prostate Cancer Early Detection. Version 1.2006. National Comprehensive Cancer Network, 2006. Available online. Last accessed March 5, 2007. 
    
    
17. Pepe MS, Etzioni R, Feng Z, et al.: Phases of biomarker development for early detection of cancer. J Natl Cancer Inst 93 (14): 1054-61, 2001.  [PUBMED Abstract]
    
    
18. Woolf SH: Screening for prostate cancer with prostate-specific antigen. An examination of the evidence. N Engl J Med 333 (21): 1401-5, 1995.  [PUBMED Abstract]
    
    
19. Grosse SD, Khoury MJ: What is the clinical utility of genetic testing? Genet Med 8 (7): 448-50, 2006.  [PUBMED Abstract]
    
    
20. Zheng SL, Sun J, Wiklund F, et al.: Cumulative association of five genetic variants with prostate cancer. N Engl J Med 358 (9): 910-9, 2008.  [PUBMED Abstract]
    
    
21. Sun J, Chang BL, Isaacs SD, et al.: Cumulative effect of five genetic variants on prostate cancer risk in multiple study populations. Prostate 68 (12): 1257-62, 2008.  [PUBMED Abstract]
    
    
22. Bauer JJ, Srivastava S, Connelly RR, et al.: Significance of familial history of prostate cancer to traditional prognostic variables, genetic biomarkers, and recurrence after radical prostatectomy. Urology 51 (6): 970-6, 1998.  [PUBMED Abstract]
    
    
23. Bova GS, Partin AW, Isaacs SD, et al.: Biological aggressiveness of hereditary prostate cancer: long-term evaluation following radical prostatectomy. J Urol 160 (3 Pt 1): 660-3, 1998.  [PUBMED Abstract]
    
    
24. Spangler E, Zeigler-Johnson CM, Malkowicz SB, et al.: Association of prostate cancer family history with histopathological and clinical characteristics of prostate tumors. Int J Cancer 113 (3): 471-4, 2005.  [PUBMED Abstract]
    
    
25. Kupelian PA, Kupelian VA, Witte JS, et al.: Family history of prostate cancer in patients with localized prostate cancer: an independent predictor of treatment outcome. J Clin Oncol 15 (4): 1478-80, 1997.  [PUBMED Abstract]
    
    
26. Kupelian PA, Reddy CA, Reuther AM, et al.: Aggressiveness of familial prostate cancer. J Clin Oncol 24 (21): 3445-50, 2006.  [PUBMED Abstract]
    
    

Back to Top

< Previous Section  |  Next Section >