Abstract
PURPOSE
Prostate biopsy is often recommended based on elevations in prostate-specific antigen (PSA) and/or abnormal digital rectal examination (DRE). We investigate the stability of a single positive test over the next 3 consecutive years.
METHODS
2578 participants in a San Antonio screening cohort with two or more consecutive annual PSA and DRE tests were identified. Numbers of occurrences of an elevated PSA (≥ 2.5 ng/ml) followed by one or more non-elevated PSA results were compared with similar fluctuations of DRE from abnormal to normal.
RESULTS
Among 2272 men who never had a biopsy performed during the study, in 23.3% of 744 incidences of an elevated PSA with one year of follow-up, the next PSA was not elevated, in 19.5% of 353 incidences of an elevated PSA with two years of follow-up, the next two consecutive PSAs were not elevated, and in 17.5% of 68 incidences of an elevated PSA with three years of follow-up, the next three consecutive PSAs were not elevated. Rates were similar but lower among 221 men with one or more negative biopsies during the study and 85 men who eventually developed prostate cancer during the study. In contrast, approximately 70% of abnormal DREs were normal the following year, even among prostate cancer cases, and in the majority of incidences, remained normal the next 2 to 3 consecutive years.
CONCLUSIONS
Occurrences of reversed PSA cutpoint- or abnormal DRE-based decisions to biopsy one or more years after the initial test are not uncommon, suggesting repetition of these tests.
Keywords: Prostate-specific antigen, digital rectal exam, yearly fluctuation
INTRODUCTION
Prostate-specific antigen (PSA) levels in serum and digital rectal examination (DRE) are the most commonly used tools for screening for prostate cancer (1). The widespread use of PSA and DRE have resulted in a dramatic rise in prostate cancer detection over the past decade (2). While the rate of detection of organ-confined disease has increased with subsequent improvement in curability, there has been an undisputed rise in the detection of clinically insignificant cancers subjected to unnecessary treatment and related morbidity (3). The use of PSA and DRE for screening in prostate cancer detection has been a subject of relentless debate (4). Outcomes regarding potential benefits of screening will not be known until the results of the ongoing Prostate, Lung, Colorectal and Ovarian Cancer trial in the United States and the European Randomized Study of Screening for Prostate Cancer in Europe are reported (5,6). Both screening tests have been used in a dichotomous manner in clinical practice; patients are generally recommended to undergo prostate biopsy if PSA exceeds a cut-off value of 2.5 or 4.0 ng/ml or if DRE is abnormal. Recent reports have suggested caution while recommending prostate biopsy based on a single value of elevated PSA (7). While natural biologic variation occurs in PSA testing in the short term, year-to-year fluctuations in the serum PSA levels have also been reported in unscreened populations (7,8). Transrectal ultrasound guided biopsy of the prostate, though technically simple, is an invasive procedure that may lead to significant bleeding and infection related complications (9). The aim of this study was to evaluate the year-to-year changes in serum PSA and DRE findings in a prospectively-studied cohort based on biopsy recommendations and biopsy findings.
MATERIALS AND METHODS
The San Antonio center of Biomarkers Of Risk for prostate cancer (SABOR) is a Clinical and Epidemiological Validation Center of the Early Detection Research Network of the National Cancer Institute. Since 2000, SABOR has recruited 3651 men without a diagnosis of prostate cancer into a longitudinal follow-up study. Participants have been followed annually by their SABOR affiliated clinician with digital rectal examination (DRE) and PSA measurement. The SABOR clinicians are credentialed to perform the DRE and are approved by the IRB to perform it. They perform the DRE blinded to the PSA result, since the serum for PSA testing is only sampled at the visit and determined in a central laboratory later. Following each annual visit a prostate biopsy is offered for a PSA level of 2.5 ng/mL or more, an abnormal DRE suspicious for prostate cancer, or, in some circumstances, a positive family history of prostate cancer.
SABOR participants with two or more annual PSA measurements available were stratified into three groups according to whether or not they had a prostate biopsy performed during the follow-up: no biopsy, one or more negative biopsies, or prostate cancer diagnosis. The latter two groups were kept distinct. The prostate cancer group included those that had prior negative biopsies, but this number was small (6 had one prior biopsy, 2 had two). Only PSA values before the prostate cancer diagnosis were included in the analysis. Nonparametric Wilcox tests were used for pair-wise comparison of continuous outcomes and covariates between the three groups. For categorical summaries the chi-square test was used.
All statistical tests were at the α=0.05 (two-sided) level of statistical significance and all statistical analyses were performed using the R statistical package (Version 2.6.0, Copyright 2007, R Foundation for Statistical Computing).
RESULTS
Of 3095 men under annual follow-up in SABOR, 2578 had two or more PSA values and were eligible for analysis. Study participants were assigned to one of three categories: those without prostate biopsy during follow-up (n=2272, 88.1%), those with one or more prostate biopsies, all negative for prostate cancer (n=221, 8.6%), and those with biopsy-detected prostate cancer (n=85, 3.3%); Table 1. Participants without prostate biopsy were statistically significantly younger than either those that had a prior negative biopsy or those who were diagnosed with prostate cancer (both p-values < 0.0001), had lower rates of a family history of prostate cancer than participants with a prior negative biopsy or cancer diagnosis (both p < 0.0001), and were more ethnically mixed than participants with a prior negative biopsy (p = 0.01). There were no statistically significant differences in any of these characteristics between participants with prior negative biopsies and those with a prostate cancer diagnosis. All three groups differed significantly from each other in terms of the cumulative number of years of follow-up (including number of annual PSA measurements available); cancer cases had the least cumulative follow-up (maximum of 5 years), followed by the group with no biopsy and the group with one or more prior biopsies (all p < 0.0001). The last two groups had a substantial fraction of subjects followed over 5 years. All three groups differed significantly from each other in terms of number of abnormal tests (DRE or PSA ≥ 2.5) per person with the prior negative biopsy and prostate cancer groups having a greater number of abnormal tests (all p < 0.02).
Table 1.
Participant characteristics
| No biopsy N=2272 |
One or more negative biopsies N=221 |
Prostate cancer cases* N=85 |
|
|---|---|---|---|
| Median (range) | |||
| Age at entry | 56.0 (28.3, 88.6) | 62.0 (33.8, 82.2) | 62.8 (45.9, 80.6) |
| Race/Ethnicity | N participants (%) | ||
| Caucasian | 1213 (53.4%) | 144 (65.2%) | 50 (58.8%) |
| African American | 285 (12.5%) | 22 (10.0%) | 13 (15.3%) |
| Hispanic | 759 (33.4%) | 54 (24.4%) | 22 (25.9%) |
| Other** | 15 (0.7%) | 1 (0.5%) | 0 (0.0%) |
| First degree family members with prostate cancer | N participants (%) | ||
| No | 1866 (82.1%) | 150 (67.9%) | 54 (63.5%) |
| Yes | 406 (17.9%) | 71 (32.1%) | 31 (36.5%) |
| Number of PSA values during follow-up | N Participants (%) | ||
| 2 | 555 (24.4%) | 17 (7.7%) | 28 (32.9%) |
| 3 | 469 (20.6%) | 33 (14.9%) | 27 (31.8%) |
| 4 | 345 (15.2%) | 25 (11.3%) | 22 (25.9%) |
| 5 | 312 (13.7%) | 44 (19.9%) | 8 (9.4%) |
| ≥ 6 | 591 (26.0%) | 102 (46.2%) | 0 (0.0%) |
| Number of abnormal DREs during follow-up | N participants (%) | ||
| 0 abnormal | 1895 (83.4%) | 114 (51.6%) | 58 (68.2%) |
| 1 | 145 (6.4%) | 40 (18.1%) | 15 (17.6%) |
| ≥ 2 | 41 (1.8%) | 36 (16.3%) | 5 (5.9%) |
| Not performed or result unknown | 191 (8.4%) | 31 (14.0%) | 7 (8.2%) |
| Number of PSAs ≥ 2.5 ng/mL during follow-up | N participants (%) | ||
| 0 | 1806 (79.5%) | 94 (42.5%) | 29 (34.1%) |
| 1 | 195 (8.6%) | 29 (13.1%) | 18 (21.2%) |
| 2 | 107 (4.7%) | 17 (7.7%) | 19 (22.4%) |
| 3 | 67 (2.9%) | 28 (12.7%) | 13 (25.3%) |
| 4 | 43 (1.9%) | 15 (6.8%) | 4 (4.7%) |
| 5 | 31 (1.4%) | 10 (4.5%) | 2 (2.4%) |
| ≥ 6 | 23 (1.0%) | 28 (12.7%) | 0 (0.0%) |
6 prostate cancer cases had one prior negative biopsy, 3 had two prior negative biopsies
American Indian, Asian or Hawaiian/pacific islander
Observed fluctuations of the PSA test after an elevation greater than or equal to 2.5 ng/mL is shown in Table 2. In the majority of incidences an elevated PSA was followed by consecutive elevated PSAs one, two, and even three consecutive years later. However, in not insignificant fractions of cases, the next consecutive annual PSAs reverted to “not elevated” and remained there. Among men who never had a biopsy performed during the study, in 23.3% of incidences the next PSA was not elevated, in 19.5% the next two consecutive PSAs were not elevated and in 17.5% the next three consecutive PSAs were not elevated. Notably the tendency of an elevated PSA to revert to “not elevated” at the next visit did not substantially decrease with higher PSA values. In men with no biopsy performed, for elevated PSA levels in the ranges 2.5 to 4.0 ng/mL, 4.0 to 6.0 ng/mL, 6.0 to 10.0 ng/mL and > 10.0 ng/mL, fractions of time the next annual PSA was not elevated were 27.9%, 14.6%, 9.7% and 26.7%, respectively (n=499, 158, 72, 15, respectively). Persistence of an elevated PSA over the ensuing one to three years more commonly occurred in men with one or more negative biopsies performed during study or with an eventual prostate cancer diagnosis (p ≤ 0.03 for comparisons of these groups to the no biopsy group). The median (range) elevated PSA value among the no biopsy, one or more negative biopsy and prostate cancer groups were 3.3 (2.5, 29.8), 4.1 (2.5, 19.5) and 3.2 (2.5, 8.9) ng/mL, respectively. PSA decreased by a median 3.1% at the next annual visit in men with no biopsy, and by a median 3.6% in men who had one or more negative biopsies performed. In contrast, among men eventually diagnosed with prostate cancer, PSA increased by a median 13.6% at the next annual visit (p < 0.0002 for both comparisons to other groups).
Table 2.
PSA results after an elevation greater than or equal to 2.5 ng/mL. Elevated denotes PSA ≥ 2.5 ng/mL, not elevated PSA < 2.5 ng/mL; Entries in table are number (%) of incidences.
| No biopsy | One or more negative biopsies |
Prostate cancer case |
|
|---|---|---|---|
| Next consecutive PSA following an elevated PSA* | |||
| Not elevated | 173 (23.3%) | 61 (17.3%) | 6 (8.8%) |
| Elevated | 571 (76.7%) | 292 (82.7%) | 62 (91.2%) |
| Next 2 consecutive PSAs following an elevated PSA** | |||
| Both not elevated | 90 (19.5%) | 29 (11.1%) | 0 (0.0%) |
| Both elevated | 312 (67.5%) | 194 (74.0%) | 26 (86.7%) |
| First not elevated, second elevated | 28 (6.1%) | 14 (5.3%) | 3 (10.0%) |
| First elevated, second not elevated | 32 (6.9%) | 25 (9.5%) | 1 (3.3%) |
| Next 3 consecutive PSAs following an elevated PSA*** | |||
| All 3 not elevated | 50 (17.5%) | 17 (9.5%) | 0 (0.0%) |
| All 3 elevated | 165 (57.9%) | 117 (65.4%) | 8 (88.9%) |
| Only 1 elevated | 35 (12.3%) | 33 (18.4%) | 1 (11.1%) |
| Only 2 elevated | 35 (12.3%) | 12 (6.7%) | 0 (0.0%) |
Number of elevated PSAs [participants] with at least one subsequent PSA required to be included in analysis was 744 [368], 353 [124], and 68 [40] for the no biopsy, one or more prior biopsy and prostate cancer groups respectively.
Number of elevated PSAs [participants] with at least two subsequent PSAs required to be included in analysis was 462 [246], 262 [106], and 30 [22] for the respective groups.
Number of elevated PSAs [participants] with at least three subsequent PSAs required to be included in analysis was 285 [168], 179 [82], and 9 [7] for the respective groups.
Fluctuation of the DRE result from abnormal to normal greatly exceeded that for PSA. Approximately 70% of abnormal DREs were normal the following year (Table 3). Even among the 11 instances of an abnormal DRE with one subsequent DRE available before a prostate cancer diagnosis, in 8 (72.7%), the DRE reverted to normal the next year. An abnormal DRE was followed by two and three consecutive annual normal DREs 69.1% and 68.0% of the time, respectively, among men who had no biopsies performed during follow-up and 50.5% and 42.4% of the time, respectively, among men who had one or more negative biopsies during the course of follow-up. Even among men eventually diagnosed with prostate cancer, in the majority of instances an abnormal DRE was followed by two or three consecutive yearly normal DREs (Table 3). Combining the PSA and DRE test to indicate a positive result if either PSA was greater than or equal to 2.5 ng/mL or the DRE was abnormal did not improve stability, but rather showed similar but worse characteristics than the PSA 2.5 ng/mL criterion. For example, in the group of men with no biopsy performed in the study, in 234 of 817 (28.6%) instances of a positive combined test the test reverted to negative the next year. In the same group of men, in 121 of 505 (24.0%) instances of a positive combined test the next two annual tests were negative and in 63 of 299 (21.1%) of instances of a positive combined test (where there were at least three consecutive annual follow-up tests) the next three annual consecutive tests were all negative.
Table 3.
DRE results after an abnormal DRE finding. Entries in table are number (%) of incidences.
| No biopsy | One or more negative biopsies |
Prostate cancer case |
|
|---|---|---|---|
| Next DRE result following an abnormal DRE* | |||
| Normal | 115 (69.7%) | 76 (67.3%) | 8 (72.7%) |
| Abnormal | 146 (30.3%) | 37 (32.7%) | 3 (27.3%) |
| Next 2 DRE results following an abnormal DRE** | |||
| Both normal | 67 (69.1%) | 46 (50.5%) | 5 (62.5%) |
| Both abnormal | 12 (12.4%) | 8 (8.8%) | 0 (0.0%) |
| First normal, Second abnormal | 7 (7.2%) | 12 (13.2%) | 2 (25.0%) |
| First abnormal, Second normal | 11 (11.3%) | 25 (27.5%) | 1 (12.5%) |
| Next 3 DRE results following an abnormal DRE*** | |||
| All 3 normal | 34 (68.0%) | 25 (42.4%) | 3 (60.0%) |
| All 3 abnormal | 2 (4.0%) | 1 (1.7%) | 0 (0.0%) |
| One abnormal | 2 (4.0%) | 11 (18.6%) | 1 (20.0%) |
| Two abnormal | 12 (24.0%) | 22 (37.3%) | 1 (20.0%) |
Number of abnormal DRE findings [participants] with at least one subsequent DRE test performed as required to be included in analysis was 165 [138], 113 [75], and 11 [10] for the no biopsy, one or more prior biopsy and prostate cancer groups respectively.
Number of abnormal DRE findings [participants] with at least two subsequent DRE tests performed as required to be included in analysis was 97 [83], 91 [62], and 8 [7] for the three respective groups.
Number of abnormal DRE findings [participants] with at least three subsequent DRE tests performed as required to be included in analysis was 50 [46], 59 [39], and 5 [5] for the three respective groups.
DISCUSSION
Serum PSA and DRE are the most commonly used tools used for prostate cancer screening. Results from this study highlight the flaws of these screening approaches that use tests which are highly variable. Previous studies have evaluated the year-to-year variability of PSA (7). Eastham et al evaluated 972 participants in the Polyp Prevention trial, all without prostate biopsy and with unknown prostate cancer status, followed for 4 years as part of a study that did not screen for prostate cancer. Outcome measures included an abnormal test based on a PSA level higher than 4 ng/mL, a PSA level higher than 2.5 ng/mL, a PSA level above the age-specific cutoff, a PSA level in the range of 4 to 10 ng/mL and free-to-total PSA ratio of less than 0.25 ng/mL, and a PSA velocity higher than 0.75 ng/mL per year. They found that prostate biopsy would have been recommended in 21% participants with a PSA level above 4 ng/mL and in 37% with a level higher than 2.5 ng/mL. Among men with an abnormal PSA finding, a high proportion had a normal PSA at one or more subsequent visits during 4 years of follow-up: 44% of participants with a PSA level above 4 ng/mL and 40% with a level above 2.5 ng/mL. Based on these findings, the authors concluded that an elevated PSA should be confirmed several weeks later before proceeding with prostate biopsy. A limitation of their study was the lack of biopsy data in men who developed an elevated PSA level. In contrast, the current study evaluated the roles of both PSA as well as DRE in a prospective cohort in which subjects with elevated PSA and/or DRE underwent prostate biopsy. Despite the inherent differences in these patient populations, the findings are strikingly similar. In our study, approximately 23% and 40% of men in the no biopsy group dropped below a PSA value of 2.5 and 4.0 ng/ml, respectively, the next year after the elevated PSA compared to 26% and 30% in the Eastham study. With each additional year of follow-up, although the rate of PSA falling below the cut-off level decreased, it remained significant from 23.3% at one year to19.5% at 2 years and 17.5 % at 3 years in the no biopsy group. The variability in the DRE results on a yearly basis in this study confirms previous reports of the unreliability and inter-observer variability of DRE as a screening tool for prostate cancer (10–12). Adding DRE to PSA decreased the performance for prostate cancer screening compared to PSA alone while improving it compared to DRE alone.
These data confirm that PSA and DRE, alone or in combination, have significant annual variability making it difficult to justify their role as the only recommended screening tools for prostate cancer. The unreliability of PSA and DRE testing was seen even in patients diagnosed with prostate cancer, a finding in agreement with observations from prior studies of active surveillance for localized prostate cancer (13, 14).
Our results have implications for the way PSA and DRE are viewed as screening tools for prostate cancer. Based on these results, it is reasonable to repeat both PSA and DRE tests before recommending prostate biopsy. The timing of the repeat test would be a flexible decision between clinician and patient. If there was significant concern, it could be done earlier, perhaps within a few weeks. If there was less concern, with less-elevated PSA levels, a less concerning DRE or an abnormal DRE accompanying a low PSA, it might be reasonable to repeat the test in 6–12 months. Performing confirmatory PSA and DRE tests would most likely improve specificity of prostate cancer screening. Conversely, data from the Prostate Cancer Prevention Trial have shown that PSA is a continuous and not a dichotomous marker for prostate cancer detection (15,16). As such, even if the PSA is repeated and falls below the cut off levels of 2.5 or 4 ng/ml, there is still a risk of prostate cancer and of high grade disease. One approach that would improve the current screening paradigm is to use a risk-based, individualized approach to prostate cancer detection via risk calculators and nomograms which estimate a person’s risk of biopsy-detectable prostate cancer incorporating PSA and DRE, augmenting their predictive utility by including age, race, family history of prostate cancer, history of a prior negative biopsy, the presence of lower urinary tract voiding symptoms, free: total PSA ratio, and new biomarkers for prostate cancer as they are discovered and validated (17,18).
A limitation regarding the results concerning PSA and DRE fluctuation in the group of men eventually diagnosed with prostate cancer is that they only pertain to the period before the prostate cancer diagnosis. In the SABOR screening study follow-up ceased at the time of diagnosis so there are no data to inform on fluctuation of PSA and DRE after diagnosis. Since many screen-detected cancers are likely to have a long latent period and not be immediately treated, it would be of interest to measure expected fluctuation in this circumstance. Long-term follow-up of a surveillance cohort will be useful for this purpose.
CONCLUSIONS
A single PSA or DRE finding is an unreliable measure for subjecting patients to prostate biopsy. Although more stable than the DRE, occurrences of reversed PSA-cutpoint-based decisions to biopsy one or more years later are not uncommon, supporting previous suggestions that elevated PSA tests be repeated. The inherent deficiencies of single results from screening tools for prostate cancer should be kept in mind while counseling patients concerning prostate cancer detection.
Acknowledgments
Funding/Support: This study was supported in part by the following US Public Health Service grants awarded by the National Cancer Institute, Department of Health and Human Services: CA 37429, CA 35178, CA 45808, CA 86402, and 5UO1CA86402 (Early Detection Research Network, National Cancer Institute, U.S. National Institutes of Health)
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
REFERENCES
- 1.Catalona WJ, Richie JP, Ahmann FR, Hudson MA, Scardino PT, Flanigan RC, et al. Comparison of digital rectal examination and serum prostate specific antigen in the early detection of prostate cancer: results of a multicenter clinical trial of 6,630 men. J Urol. 1994;151:1283. doi: 10.1016/s0022-5347(17)35233-3. [DOI] [PubMed] [Google Scholar]
- 2.Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, et al. Cancer statistics, 2008. CA Cancer J Clin. 2008;58:71. doi: 10.3322/CA.2007.0010. [DOI] [PubMed] [Google Scholar]
- 3.Epstein JI, Walsh PC, Brendler CB. Radical prostatectomy for impalpable prostate cancer: the Johns Hopkins experience with tumors found on transurethral resection (stages T1A and T1B) and on needle biopsy (stage T1C) J Urol. 1994;152:1721. doi: 10.1016/s0022-5347(17)32370-4. [DOI] [PubMed] [Google Scholar]
- 4.Crawford ED, Thompson IM. Controversies regarding screening for prostate cancer. BJU Intl. 2007;100(Suppl 2):5. doi: 10.1111/j.1464-410X.2007.06943.x. [DOI] [PubMed] [Google Scholar]
- 5.Andriole GL, Levin DL, Crawford ED, Gelmann EP, Pinsky PF, Chia D, et al. Prostate Cancer Screening in the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial: findings from the initial screening round of a randomized trial. JNCI. 2005;97:433. doi: 10.1093/jnci/dji065. [DOI] [PubMed] [Google Scholar]
- 6.Roobol MJ, Schroder FH. European Randomized Study of Screening for Prostate Cancer: achievements and presentation. BJU Intl. 2003;92(Suppl 2):117. doi: 10.1111/j.1464-410x.2003.4698x.x. [DOI] [PubMed] [Google Scholar]
- 7.Eastham JA, Riedel E, Scardino PT, Shike M, Fleisher M, Schatzkin A, et al. Variation of serum prostate-specific antigen levels: an evaluation of year-to-year fluctuations. JAMA. 2003;289:2695. doi: 10.1001/jama.289.20.2695. [DOI] [PubMed] [Google Scholar]
- 8.Nixon RG, Wener MH, Smith KM, Parson RE, Strobel SA, Brawer MK. Biological variation of prostate specific antigen levels in serum: an evaluation of day-to-day physiological fluctuations in a well-defined cohort of 24 patients. J Urol. 1997;157:2183. doi: 10.1016/s0022-5347(01)64711-6. [DOI] [PubMed] [Google Scholar]
- 9.Rodriguez LV, Terris MK. Risks and complications of transrectal ultrasound guided prostate needle biopsy: a prospective study and review of the literature. J Urol. 1998;160:2115. doi: 10.1097/00005392-199812010-00045. [DOI] [PubMed] [Google Scholar]
- 10.Gosselaar C, Kranse R, Roobol MJ, Roemeling S, Schroeder FH. The interobserver variability of digital rectal examination in a large randomized trial for the screening of prostate cancer. Prostate. 2008;68:985. doi: 10.1002/pros.20759. [DOI] [PubMed] [Google Scholar]
- 11.Gosselaar C, Roobol MJ, Roemeling S, Schroeder FH. The Role of the Digital Rectal Examination in Subsequent Screening Visits in the European Randomized Study of Screening for Prostate Cancer (ERSPC), Rotterdam. European Urol. 2008;54:581. doi: 10.1016/j.eururo.2008.03.104. [DOI] [PubMed] [Google Scholar]
- 12.Gosselaar C, Roobol MJ, van den Bergh RC, Wolters T, Schroeder FH. Digital Rectal Examination and the Diagnosis of Prostate Cancer-a Study Based on 8 Years and Three Screenings within the European Randomized Study of Screening for Prostate Cancer (ERSPC), Rotterdam. European Urol. 2008 doi: 10.1016/j.eururo.2008.03.079. epub ahead of print. [DOI] [PubMed] [Google Scholar]
- 13.Roemeling S, Roobol MJ, de Vries SH, Wolters T, Gosselaar C, van Leenders GH, et al. Active surveillance for prostate cancers detected in three subsequent rounds of a screening trial: characteristics, PSA doubling times, and outcome. European Urol. 2007;51:1244. doi: 10.1016/j.eururo.2006.11.053. discussion 51. [DOI] [PubMed] [Google Scholar]
- 14.Klotz L. Active surveillance for favorable risk prostate cancer: rationale, risks, and results. Urologic oncology. 2007;25:505. doi: 10.1016/j.urolonc.2007.05.021. [DOI] [PubMed] [Google Scholar]
- 15.Thompson IM, Goodman PJ, Tangen CM, Lucia MS, Miller GJ, Ford LG, et al. The influence of finasteride on the development of prostate cancer. N Engl J Med. 2003;349:215. doi: 10.1056/NEJMoa030660. [DOI] [PubMed] [Google Scholar]
- 16.Thompson IM, Pauler DK, Goodman PJ, Tangen CM, Lucia MS, Parnes HL, et al. Prevalence of prostate cancer among men with a prostate-specific antigen level < or=4.0 ng per milliliter. N Engl J Med. 2004;350:2239. doi: 10.1056/NEJMoa031918. [DOI] [PubMed] [Google Scholar]
- 17.Thompson IM, Ankerst DP, Chi C, Goodman PJ, Tangen CM, Lucia MS, et al. Assessing prostate cancer risk: results from the Prostate Cancer Prevention Trial. JNCI. 2006;98:529. doi: 10.1093/jnci/djj131. [DOI] [PubMed] [Google Scholar]
- 18.Nam RK, Toi A, Klotz LH, Trachtenberg J, Jewett MA, Appu S, et al. Assessing individual risk for prostate cancer. J Clin Oncol. 2007;25:3582. doi: 10.1200/JCO.2007.10.6450. [DOI] [PubMed] [Google Scholar]