This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mäkinen, T. Articles by Auvinen, A. Search for Related Content PubMed PubMed Citation Articles by Mäkinen, T. Articles by Auvinen, A. Social Bookmarking                            What's this?

Family History and Prostate Cancer Screening With Prostate-Specific Antigen

From the Finnish Cancer Registry, and the Departments of Clinical Chemistry and Urology, Helsinki University Central Hospital, Helsinki; Department of Surgery, Jorvi Hospital, Espoo; Department of Surgery, Seinäjoki Central Hospital, Seinäjoki; and Department of Urology, Tampere University Hospital, and Medical School and School of Public Health, University of Tampere, Tampere, Finland.

Address reprint requests to Tuukka Mäkinen, MD, Finnish Cancer Registry, Liisankatu 21B, FIN-00170 Helsinki, Finland; email: tuukka.makinen{at}cancer.fi

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: Early detection of prostate cancer has been recommended for men with affected first-degree relatives despite the lack of evidence for mortality reduction. We therefore evaluated the impact of family history in the Finnish prostate cancer screening trial.

PATIENTS AND METHODS: Approximately 80,000 men were identified from the population register for the first screening round. Of the 32,000 men randomized to the screening arm, 30,403 were eligible at the time of invitation. A blood sample was drawn from the participants (n = 20,716), and serum prostate-specific antigen (PSA) was determined. Men with a PSA level 4.0 ng/mL were referred for prostate biopsy. Information on family history was obtained through a self-administered questionnaire at baseline.

RESULTS: A total of 964 (5%) of the 20,716 screening participants had a positive family history, and 105 (11%) were screening-positive. Twenty-nine tumors were diagnosed, corresponding to a detection rate of 3.0% (29 of 964) and a positive predictive value of 28% (29 of 105). Of the 19,347 men without a family history, 1,487 (8%) had a PSA level 4.0 ng/mL. The detection rate was 2.4% (462 of 19,347) and the positive predictive value was 31% (462 of 1,487). The risk associated with a positive family history was not substantially increased (rate ratio, 1.3; 95% confidence interval, 0.9 to 1.8). The results were not affected by the age of the screenee or age at diagnosis of the affected relative. The program sensitivity was 6% (29 of 491) (ie, selective screening policy would have missed 94% of cancers in the population). No differences were seen in the characteristics of screen-detected cancers by family history.

CONCLUSION: Our findings provide no support for selective screening among men with affected relatives.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
PROSTATE CANCER screening is now common practice despite the lack of evidence for mortality reduction. One of the few established risk factors for prostate cancer is a family history of the disease, particularly for men with a family history of an early-onset prostate cancer.¹ In the United States, prostate-specific antigen (PSA)–based screening has been recommended by the American Urological Association and the American Cancer Society, especially for men with affected first-degree relatives.^2,3 Little is known, however, about the impact of PSA screening among men with a family history of prostate cancer. Selective screening of subgroups of the population with an increased risk of prostate cancer may improve program performance, that is, increase the detection rate in the high-risk population and program specificity in the target population (effectively identify men free of cancer), but has the disadvantage of low program sensitivity (only a small proportion of cancers in the target population are detected).⁴

In the first round of the Finnish prostate cancer screening trial (1996 through 1999), we compared the process measures of screening (ie, detection rate, positive predictive value, and specificity of PSA testing) between men with and without a family history. A family history was defined as positive if any first-degree relative was affected; however, the information of second-degree relatives was also collected. The importance of a family history at the population level was assessed in terms of program sensitivity and specificity.

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Subjects
The Finnish prostate cancer screening trial was initiated in May 1996. It is the largest component in the European Randomized Study of Screening for Prostate Cancer.⁵ Approximately 80,000 men aged 55 to 67 years (born 1929 to 1944) were enrolled from the population register during the first round in 1996 to 1999. Men with a diagnosis of prostate cancer before randomization were excluded. A total of 32,000 men were randomized onto the screening arm, and the remaining men formed the control arm. Persons who had died or moved outside the study area by the time of invitation or had forbidden the use of their addresses in the national population database for any purpose were excluded. At the time of invitation (an average of 6 months after randomization), 30,403 men were eligible for screening and 20,716 (68%) eventually participated. Our report is derived from cross-sectional data collected within a randomized screening trial. The study population was formed at baseline, with exposure contrast defined on the basis of family history.

Baseline Questionnaire
Information on family history was obtained by means of a questionnaire at the time of invitation. If a subject reported one or more first-degree relatives (ie, father or brother) diagnosed with prostate cancer, the family history was regarded as positive. Because of confidentiality reasons, we were not able to identify the affected relatives or confirm reported diagnoses from medical records.

Laboratory Methods
After informed consent, a blood sample was drawn. The serum concentration of total PSA was determined by the Hybritech Tandem-E assay (Beckman Coulter, San Diego, CA). All analyses were carried out at the Department of Clinical Chemistry, Helsinki University Central Hospital.

Screening Algorithm
All men with a PSA level 4.0 ng/mL were referred for diagnostic examinations involving digital rectal examination (DRE), transrectal ultrasound (TRUS) examination, and prostate sextant biopsies, supplemented by a directed biopsy if a focal finding was detected by either DRE or TRUS examination. DRE findings were regarded as suspicious if nodularity, induration, or asymmetry was found. In TRUS, a hypoechoic lesion was regarded as suggestive of prostate cancer.

Diagnostics
All diagnoses were made on the basis of histologic examination. Clinical staging was conducted according to the tumor, node, metastasis system.⁶ Histologic characteristics at biopsy were graded according to the Gleason score system.⁷ All cancers detected as a result of screening were regarded as screen-detected, regardless of the time interval from the initial PSA test.

Data Analyses
The ratio of detection rates (rate ratio [RR]) with 95% confidence interval (CI) was calculated for men with an affected family member(s) relative to those with no such family history. The 405 men with family history missing were excluded from all the analyses. The risk of prostate cancer was analyzed separately for screenees below and above 60 years of age, as well as by the age of an affected relative at diagnosis. The age of an affected relative was unavailable for 74 (0.4%) men who were excluded from this analyses. None of them were diagnosed with prostate cancer. Student’s t test was used for the comparison of the mean ages at diagnosis by family history. For comparison of tumor characteristics, patients with unavailable Gleason score at biopsy (n = 5) or unavailable clinical stage (n = 3) were excluded. PSA concentrations by family history were compared by the Wilcoxon signed rank test. The specificity of the PSA test is given separately for men with and without family history, and corresponds to the proportion of men with PSA levels less than 4 ng/mL among all healthy screening participants with the corresponding family history. The approximation of program sensitivity for family history as a supplemental screening criterion indicates the proportion of cancers detectable by selective screening policy in the screened target population with lacking information on interval cancers.⁸ In other words, it corresponds to the proportion of family history–positive cases with a PSA level 4 ng/mL among all screen-detected cancers with family history available. Program specificity for family history as a supplemental screening criterion indicates the proportion of men correctly identified as free of prostate cancer (ie, healthy screenees with a negative test combination of family history and PSA) among all healthy screening participants with family history known. Statistical analyses were performed on CIA Version 1.1 (Martin J. Gardner and British Medical Journal) and S-PLUS Version 4.0 (MathSoft Inc, Cambridge MA).

Ethics
The ethical committee of each participating hospital approved the trial protocol. Permission to obtain medical records was obtained from the Ministry of Social Affairs and Health and for cancer registry data from the National Research and Development Center for Welfare and Health.

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Of the 20,716 participants, 98.0% (20,311 of 20,716) provided information regarding family history through a self-administered questionnaire before screening. A total of 964 of 20,311 (4.7%) gave a positive family history (ie, one or more affected first-degree relatives). Of these, 708 reported a father with a prostate cancer, and 266 had an affected brother (Table 1). Only 17 men had two or more affected first-degree relatives. Ninety (0.4%) men had a father or brother(s) affected at the age of 59 years or less. Approximately 300 had an affected uncle (either maternal or paternal), but few were aware of a prostate cancer diagnosed among their grandfathers. A family history of prostate cancer was associated with neither age nor screening center (data not shown).

Men reporting a positive family history did not have a substantially increased risk of prostate cancer (RR, 1.3; 95% CI, 0.9 to 1.8). No materially increased prostate cancer risk was observed for men with an affected father, or affected brother, or other affected relative on either side of the family. The effect of a positive family history was also similar in the subgroup of men aged less than 60 years (Table 1). Of those 17 men with two or more first-degree relatives affected, only one was diagnosed with prostate cancer (RR, 2.5; 95% CI, 0.1 to 14.4).

Screenees with an affected first-degree relative with an early-onset prostate cancer were not more often diagnosed with a cancer at screening than those without such history. Of the 90 men with a father or brother(s) affected before the age of 60, only eight were screening-positive. Three of them were diagnosed with cancer, giving a detection rate of 3.3% (95% CI, 0.7% to 9.4%), an RR of 1.4 (95% CI, 0.5 to 4.3), and a positive predictive value of 38% (95% CI, 8.5% to 75.5%) (Table 2).

No significant differences were seen in the characteristics of the screen-detected cancers by family history. The mean age at diagnosis was 61 years among men both with and without a family history (P = .62). The PSA concentrations of tumors detected were also comparable, with median values of 6.2 ng/mL and 7.5 ng/mL, respectively (P = .24). Family history was not associated with Gleason score. The detection rate of clinically organ-confined cancers was 2.7% (95% CI, 1.8% to 3.9%; 26 of 964) among men with and 2.0% (95% CI, 1.9% to 2.3%; 396 of 19,347) among those without a family history (Table 3).

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

Controversy seems to prevail as to whether men with a family history have tumors presenting with features usually associated with a more aggressive type of disease.^11,12 In our trial, family history was not associated with prognostic indicators such as Gleason score, clinical stage, PSA, or age at diagnosis, and hence suggested no greater importance of early detection of prostate cancer among men with a family history. However, we cannot exclude the possibility that these tumors would develop differently if they were detected years later on the basis of clinical symptoms.

The present findings are inconsistent with those in previous studies showing an association between a positive family history and risk of prostate cancer in population-based PSA screening. In the Quebec screening trial, a significantly elevated risk of prostate cancer was observed among men with an affected first-degree relative (RR, 1.7; 95% CI, 1.2 to 2.4).¹³ The highest risk (RR, 2.6; 95% CI, 1.7 to 4.1) was noted among men with affected brother(s). In that trial, biopsies only on men with a positive family history would have detected 14% of all tumors. The most important limitation of the Quebec study was the low attendance rate (27%), which limits the applicability of the results and increases the possibility of selection bias. Moreover, the screening algorithm was somewhat different from ours, as all men with a PSA level 3.0 ng/mL or with an abnormal DRE were referred for TRUS examination, but biopsy specimens were taken only if a hypoechoic lesion was found. Eventually, biopsies were performed on approximately half of the 1,563 screen-positive men, resulting in the diagnosis of 264 tumors (detection rate of 4.1%). The results from the Rotterdam screening trial also suggested an approximately two-fold risk associated with a family history, but again the small data set, biopsy specimens from only 202 men, limited conclusions.¹⁴

Screening with PSA has also been studied separately in high-risk families.^15,16 These studies suggested that the detection rate is relatively high among men with a family history, but provided no comparison with men without a family history or with the general population. Because the focus has thus been on small and highly selected subgroups of the population, recommendation of a more aggressive screening strategy among men with a family history, especially without evidence of mortality reduction, is hardly justified at the population level.

Figures indicating an increased risk associated with a positive family history are, however, invalid indicators of the feasibility of a selective screening program. Screening in high-risk families does not affect the validity of the screening test itself (eg, PSA) in terms of test specificity and sensitivity. Instead, the aim of selective screening is to improve the program specificity (ie, to reduce the costs and adverse effects of screening) without compromising the program sensitivity.⁴ In our trial, the program specificity of 99.6% determined on the basis of family history as a supplementary screening criterion would have eliminated nearly all the unnecessary examinations among the men free of prostate cancer. The program sensitivity of only 6% is, however, unlikely to provide substantial reduction in prostate cancer mortality at the population level.

Our trial is population-based and has a relatively high participation rate (68%), which enhances the representativeness of the material. Nearly all (98%) the screening participants provided information on family history, minimizing the possibility of selection bias. Because the information was obtained at the time of invitation (ie, before diagnostic examinations), the effect of possible diagnosis of prostate cancer on valid reporting (recall bias) was eliminated. Family history was self-reported, and could not be confirmed from medical records or the Finnish Cancer Registry. However, it has been shown that men are able to provide a family history of prostate cancer fairly accurately and reliably.¹⁷

Cross-sectional data, like ours, may be affected by two major shortcomings. First, the temporality between an exposure and an outcome remains unclear. Second, sampling of prevalent cases means selection is conditional on being diagnosed with the disease and surviving with it, which leads to inability to distinguish between risk factors for disease incidence and surviving with the disease. Our study is free from both limitations: first, exposure status (family history) was assessed before and irrespective of the screening result and diagnostic confirmation; second, the outcome measure in our article is the detection rate, which does not include prevalent, but incident cases.

PSA testing detects tumors at an early stage, which may weaken the effect of family history through overdiagnosis (ie, detection of tumors that previously remained undetected).¹⁸ Thus, the impact of a family history on detection by PSA screening may be lower than in earlier studies derived from clinically detected cases.^1,19 The extent of possible overdiagnosis cannot be evaluated, because the natural history of screen-detected tumors is not sufficiently well known. No reliable method is yet available to identify indolent tumors. Nevertheless, this would explain the discrepancy between earlier findings and ours only if the majority of the screen-detected cancers were the result of overdiagnosis, especially among the men without a family history.

Hereditary factors have been estimated to contribute to 5% to 10% or more recently up to 42% of incident cases of prostate cancer.^20,21 In our material, the corresponding population-attributable risk was 1.2%. Identification of families with a strong hereditary component demonstrates the existence of genetic factors, but provides little information regarding their importance at the population level. In other words, a rare genetic variant may have high penetrance although accounting for only a minor proportion of the cancers in the population. This has in fact been shown for prostate cancer in Finland.²² Hence, identification of a high-risk group would not justify adoption of a selective screening policy (even if the effectiveness of screening was shown). Differences in genetic factors between populations have been reported: HPCX has been suggested to account for almost half of cases with a hereditary susceptibility in Finland, whereas in North America, HPC1 seems to be the most important locus. It would therefore seem unlikely that the lack of an association between family history and screen-detected prostate cancer could be because of the absence of predisposing genes in the Finnish population.

A great deal of the evidence for an association of family history with a risk of prostate cancer is derived from clinically detected cases from the era before widespread PSA testing, but the association has also been shown more recently.^23,24 Contrary to these earlier findings, our screening program with PSA showed only a somewhat but not significantly increased risk of prostate cancer associated with a positive family history. A more important finding, however, is the poor program sensitivity of selective PSA screening on the basis of family history. Genetic factors are thus unlikely to provide a successful approach for selective prostate cancer screening. Instead of focusing on family history, the aim should be for wide coverage of the population. Currently, the cause of prostate cancer is not well enough known to enable identification of a high-risk group with a substantial population-attributable risk. In conclusion, our findings provide no support for selective screening among men with a positive family history.

	ACKNOWLEDGMENTS

 
Supported by the Academy of Finland, the Cancer Society of Finland, Helsingin Sanomat Centenarian Fund, the Research Fund of Tampere University Hospital, the Europe Against Cancer program, and the Beckman-Hybritech Corporation. In addition, T. Mäkinen received support from the Cancer Society of Pirkanmaa.

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
1. Pienta KJ, Esper PS: Risk factors for prostate cancer. Ann Intern Med 118: 793-803, 1993[Abstract/Free Full Text]

2. American Urological Association: Prostate-specific antigen (PSA) best practice policy. Oncology 14: 267-286, 2000[Medline]

3. Smith RA, Mettlin CJ, Davis KJ, et al: American Cancer Society guidelines for the early detection of cancer. CA Cancer J Clin 50: 34-49, 2000[Abstract]

4. Hakama M: Screening, in Holland WW, Detels R, Knox G (eds): Oxford Textbook of Public Health (vol 3): Applications in Public Health. Oxford, New York, Toronto, Oxford University Press, 1991, pp 91-106

5. Rationale for randomised trials of prostate cancer screening. The International Prostate Screening Trial Evaluation Group. Eur J Cancer 35:262-271, 1999

6. Sobin LH, Wittekind CH (eds): TNM Classification of Malignant Tumours: International Union Against Cancer. New York, John Wiley & Sons, 1997

7. Gleason DF: Histologic grading of prostate cancer: A perspective. Hum Pathol 23: 273-279, 1992[CrossRef][Medline]

8. Hakama M: Selective screening by risk groups, in Prorok PC, Miller AB (eds): Screening for Cancer. International Union Against Cancer Technical Report Series (vol 78). Geneva, Switzerland, International Union Against Cancer, 1984, pp 71-80

9. Xu J, Meyers D, Freije D, et al: Evidence for a prostate cancer susceptibility locus on the X chromosome. Nat Genet 20: 175-179, 1998[CrossRef][Medline]

10. Smith JR, Freije D, Carpten JD, et al: Major susceptibility locus for prostate cancer on chromosome 1 suggested by a genome-wide search. Science 274: 1371-1374, 1996[Abstract/Free Full Text]

11. Valeri A, Azzouzi R, Drelon E, et al: Early-onset hereditary prostate cancer is not associated with specific clinical and biological features. Prostate 45: 66-71, 2000[CrossRef][Medline]

12. Keetch DW, Humphrey PA, Smith DS, et al: Clinical and pathological features of hereditary prostate cancer. J Urol 155: 1841-1843, 1996[CrossRef][Medline]

13. Narod SA, Dupont A, Cusan L, et al: The impact of family history on early detection of prostate cancer. Nat Med 2: 99-101, 1995

14. Virtanen A, Gomari M, Kranse R, et al: Estimation of prostate cancer probability by logistic regression: Free and total prostate-specific antigen, digital rectal examination, and heredity are significant variables. Clin Chem 45: 987-994, 1999[Abstract/Free Full Text]

15. 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: 1275-1279, 1999[Abstract/Free Full Text]

16. McWhorter WP, Hernandez AD, Meikle AW, et al: A screening study of prostate cancer in high risk families. J Urol 148: 826-828, 1992[Medline]

17. Zhu K, McKnight B, Stergachis A, et al: Comparison of self-report data and medical records data: Results from a case-control study on prostate cancer. Int J Epidemiol 28: 409-417, 1999[Abstract/Free Full Text]

18. Catalona WJ, Smith DS, Ratliff TL, et al: Detection of organ-confined prostate cancer is increased through prostate-specific antigen-based screening. JAMA 270: 948-954, 1993[Abstract/Free Full Text]

19. Cannon L, Bishop DT, Skolnick M, et al: Genetic epidemiology of prostate cancer in the Utah Mormon genealogy. Cancer Surv 1: 47-69, 1982

20. Carter BS, Bova GS, Beaty TH, et al: Hereditary prostate cancer: Epidemiologic and clinical features. J Urol 150: 797-802, 1993[Medline]

21. Lichtenstein P, Holm NV, Verkasalo PK, et al: Environmental and heritable factors in the causation of cancer: Analyses of cohorts of twins from Sweden, Denmark, and Finland. N Engl J Med 343: 78-85, 2000[Abstract/Free Full Text]

22. Schleutker J, Matikainen M, Smith J, et al: A genetic epidemiological study of hereditary prostate cancer (HPC) in Finland: Frequent HPCX linkage in families with late-onset disease. Clin Cancer Res 6: 4810-4815, 2000[Abstract/Free Full Text]

23. Lesko SM, Rosenberg L, Shapiro S: Family history and prostate cancer risk. Am J Epidemiol 144: 1041-1047, 1996[Abstract/Free Full Text]

24. Goldgar DE, Easton DF, Cannon-Albright LA, et al: Systematic population-based assessment of cancer risk in first-degree relatives of cancer probands. J Natl Cancer Inst 86: 1600-1608, 1994[Abstract/Free Full Text]

Submitted May 7, 2001; accepted February 13, 2002.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				A. M. D. Wolf, R. C. Wender, R. B. Etzioni, I. M. Thompson, A. V. D'Amico, R. J. Volk, D. D. Brooks, C. Dash, I. Guessous, K. Andrews, et al. American Cancer Society Guideline for the Early Detection of Prostate Cancer: Update 2010 CA Cancer J Clin, March 3, 2010; (2010) caac.20066v1. [Abstract] [Full Text]

				B. J. Wilson, N. Qureshi, P. Santaguida, J. Little, J. C. Carroll, J. Allanson, and P. Raina Systematic Review: Family History in Risk Assessment for Common Diseases Ann Intern Med, December 15, 2009; 151(12): 878 - 885. [Abstract] [Full Text] [PDF]

				B. J. Wilson, N. Qureshi, P. Santaguida, J. Little, J. C. Carroll, J. Allanson, and P. Raina Systematic Review: Family History in Risk Assessment for Common Diseases Ann Intern Med, November 2, 2009; (2009) 0000605-200912150-00177v1. [Abstract] [Full Text]

				H. J. Murff, D. R. Spigel, and S. Syngal Does This Patient Have a Family History of Cancer?: An Evidence-Based Analysis of the Accuracy of Family Cancer History JAMA, September 22, 2004; 292(12): 1480 - 1489. [Abstract] [Full Text] [PDF]

				T. Makinen, T. L. J. Tammela, U.-H. Stenman, L. Maattanen, J. Aro, H. Juusela, P. Martikainen, M. Hakama, and A. Auvinen Second Round Results of the Finnish Population-Based Prostate Cancer Screening Trial Clin. Cancer Res., April 1, 2004; 10(7): 2231 - 2236. [Abstract] [Full Text] [PDF]

				D. J. Schaid The complex genetic epidemiology of prostate cancer Hum. Mol. Genet., April 1, 2004; 13(90001): R103 - 121. [Abstract] [Full Text] [PDF]

				P. Stattin, M. Hakama, G. L. Wright Jr., and P. F. Schellhammer Correspondence re: B-L. Adam et al., Serum Protein Fingerprinting Coupled with a Pattern-matching Algorithm Distinguishes Prostate Cancer from Benign Prostate Hyperplasia and Healthy Men. Cancer Res., 62: 3609-3614, 2002 Cancer Res., May 15, 2003; 63(10): 2701 - 2702. [Full Text] [PDF]

				Is Family History Relevant in Prostate Cancer Screening? Journal Watch (General), June 28, 2002; 2002(628): 7 - 7. [Full Text]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mäkinen, T. Articles by Auvinen, A. Search for Related Content PubMed PubMed Citation Articles by Mäkinen, T. Articles by Auvinen, A. Social Bookmarking                            What's this?