The European Randomised Study of Screening for Prostate Cancer (ERSPC) applies a prostate-specific antigen (PSA) cut-off value ≥3.0
To analyse the incidence and disease-specific mortality for prostate cancer (PCa) in men with an initial PSA <3.0
Design, setting and participants
From November 1993 to December 1999, a total of 42 376 men identified from population registries in the Rotterdam region (55–74 yr of age) were randomised to an intervention or control arm. A total of 19 950 men were screened during the first screening round.
A PSA <3.0
Distribution of incidence, aggressiveness, and disease-specific mortality of PCa per PSA range was measured. Causes of death were evaluated by an independent committee, and follow-up was complete until 31 December 2008.
Results and limitations
From 1993 to 2008, 915 PCa cases were diagnosed in 15 758 men (5.8%) with an initial PSA <3.0
The risk of PCa, aggressive PCa and PCa mortality in a screening population with initial PSA <3.0
Keywords: Prostatic neoplasms, Early detection of cancer, Prostate-specific antigen, Incidence, Survival.
After first being described in 1979 , prostate-specific antigen (PSA) became available as a biomarker  and potential screening tool for the early detection of prostate cancer (PCa) . With PCa being the most frequent cancer diagnosed in men and playing an important role in cancer-related deaths worldwide  and , randomised studies set up to determine whether PSA-based screening can reduce the PCa-specific mortality  and  address an important health issue. In fact, results of the European Randomised Study of Screening for Prostate Cancer (ERSPC) have shown that PSA-based screening can decrease PCa mortality by up to 30%  and .
However, despite the value of PSA in terms of lowering PCa mortality, its use in screening is limited as a result of its lack of specificity in lower PSA ranges . Since 1997, the ERSPC has used a PSA value ≥3.0
From November 1993 to December 1999, a total of 42 376 men 55–74 yr of age identified in the Rotterdam population registry were randomised to an intervention or a control arm. During the first screening round, 19 950 men were screened (excluding men previously diagnosed with PCa), with lateralised sextant biopsies being initially recommended for an abnormal digital rectal examination (DRE) or PSA level ≥4.0
During the second screening round (November 1997 to December 2003), two side studies were carried out. The first side study (November 1997 to April 2001) provided a biopsy indication for men who doubled their PSA at the second screening in the PSA range of 1–3
Aggressive cancer was defined as clinical stage ≥T2c, PSA at diagnosis >20
The number of PCa cases, overall deaths, and disease-specific deaths classified by PSA group (<1.0
A total of 21 210 men were randomised to the screening group, of which 19 950 men were actually screened. An initial PSA value <3.0
|Characteristic||PSA (ng/ml)||Total study population|
|Subjects, No. (%)||7126 (45.2)||6156 (39.1)||2476 (15.7)||15 758 (100)|
|Median age at initial PSA test, yr||61.2||62.8||64.3||62.3|
|Charlson score, No. (%)|
|5189 (72.8)||4579 (74.4)||1820 (73.5)||11 588 (73.5)|
|1638 (23.0)||1319 (21.4)||545 (22.0)||3502 (22.2)|
|266 (3.7)||208 (3.4)||93 (3.8)||567 (3.6)|
|33 (0.5)||50 (0.8)||18 (0.7)||101 (0.6)|
|PCa, No. (per 1000 LY)||129 (1.7)||415 (6.4)||371 (14.5)||915 (5.5)|
|Aggressive PCa*, No. (per 1000 LY)||27 (0.36)||61 (0.94)||50 (1.95)||138 (0.83)|
|Median age at diagnosis, yr||69.5||68.9||68.2||68.8|
|Median time to diagnosis, yr||8.3||8.1||4.2||7.1|
|Overall deaths, No. (per 1000 LY)||1560 (20.8)||1365 (21.1)||610 (23.8)||3535 (21.4)|
|PCa deaths, No. (per 1000 LY)||3 (0.04)||11 (0.17)||9 (0.35)||23 (0.14)|
Defined as clinical stage ≥T2c, Gleason-score ≥8, PSA >20
The median baseline PSA level was 1.8 (25–75p; 1.2–2.4) for men who developed PCa and 1.0 (25–75p; 0.6–1.6) for men who did not (p
In this study population, 5.8% (915 of 15 758) of all men were diagnosed with PCa at a median of 7.1 yr after initial PSA screening. The risk of PCa diagnosis increased from 1.7 per 1000 LY in men with PSA levels <1.0 ng/ml to 14.5 per 1000 LY in men with PSA levels of 2.0–2.9 ng/ml. The cumulative hazard of being diagnosed with PCa showed a 4.0- and 10.3-fold increased risk for PSA groups 1.0–1.9 and 2.0–2.9 ng/ml, respectively, compared to PSA group <1.0 ng/ml (p
|PSA (ng/ml)||Men, No. (%)||PCa, No. (% of PSA group)||Clinical stage ≤T1c, No. (% of PCa)||Clinical stage ≥T2, No. (% of PCa)||Gleason score ≤6, No. (% of PCa)||Gleason score ≥7, No. (% of PCa)||Aggressive PCa*, No. (% of PCa)|
|<1.0||7126 (45.2)||129 (1.8)||83 (64.3)||42 (32.6)||76 (58.9)||47 (36.4)||27 (20.9)|
|1.0–1.9||6156 (39.1)||415 (6.7)||276 (66.5)||132 (31.8)||310 (74.7)||89 (21.4)||61 (14.7)|
|2.0–2.9||2476 (15.7)||371 (15.0)||243 (65.5)||122 (32.9)||279 (75.2)||73 (19.7)||50 (13.5)|
|Total||15 758 (100)||915 (5.8)||602 (65.8)||296 (32.3)||665 (72.7)||209 (22.8)||138 (15.1)|
Defined as clinical stage ≥T2c, Gleason score ≥8, PSA >20
Interval-detected PCa cases were aggressive more often (72 of 182; 39.6%) than screen-detected cases (66 of 733; 9.0%; p
|PSA (ng/ml)||PCa (95% CI)||Aggressive PCa (95% CI)||Overall mortality (95% CI)||PCa mortality (95% CI)|
* Reference group.
The number of overall deaths and disease-specific deaths classified by PSA group is outlined in Table 1. After a 15-yr period, an overall mortality rate of 21.4 men per 1000 LY was observed (3535 of 15 758; 22.4%), while PCa-specific mortality reached 0.14 per 1000 LY (23 of 15 758; 0.15%). The risk of PCa-specific mortality increased with higher initial PSA levels from 0.04 per 1000 LY in the PSA range <1.0 ng/ml to an incidence of 0.35 per 1000 LY in the range 2.0–2.9 ng/ml. PCa-specific survival stratified by PSA group and corrected for age and comorbidity is shown in Fig. 4. There was a significant increase in risk of PCa death, which was 4.0 and 7.6 times higher for the PSA groups 1.0–1.9 and 2.0–2.9 ng/ml, respectively, than for the group PSA <1.0 ng/ml. Disease-specific mortality occurred in 5 of 733 (0.7%) screen-detected cases and in 18 of 182 (9.9%) interval-detected cases (p
|PSA (ng/ml)||Men, No. (%)||PCa, No. (% of PSA group)||Aggressive PCa*, No. (% of screening- or interval-detected PCa)||PCa deaths, No. (% of screening- or interval-detected PCa)|
|<1.0||7126 (45.2)||92 (1.3)||37 (0.5)||11 (12.0)||16 (43.2)||0 (0)||3 (8.1)|
|1.0–1.9||6156 (39.1)||345 (4.8)||53 (0.7)||29 (8.4)||32 (60.4)||4 (1.2)||7 (13.2)|
|2.0–2.9||2476 (15.7)||296 (4.2)||48 (0.7)||26 (8.8)||24 (50.0)||1 (0.3)||8 (16.7)|
|Total||15 758 (100)||733 (4.7)||182 (2.6)||66 (9.0)||72 (39.6)||5 (0.7)||18 (9.9)|
Defined as clinical stage ≥T2c, Gleason-score ≥8, PSA >20
Results of the ERSPC have shown that early detection by PSA-based screening can decrease PCa mortality up to 30% after a median follow-up of 9 yr  and . The Göteborg screening study, which is part of ERSPC, recently showed even more favourable results after 14-yr follow-up . However, it is also known that PCa is prevalent in men with PSA levels below commonly used thresholds, with PSA lacking specificity, especially in these lower ranges . This study was set up to quantify the predictive value of a negative result in PSA-based screening, to evaluate the risk of postponing PCa diagnosis in screening by applying a PSA cut-off ≥3.0
We identified a significant increase in PCa incidence in men with initial PSA values of 1.0–1.9 ng/ml and 2.0–2.9 ng/ml of 4.0- and 10.3-fold, respectively, compared to men with initial PSA levels <1.0 ng/ml. Similar results were observed for aggressive PCa, which showed a 2.7- and 6.2-fold increased risk in these groups, respectively. When we examined alternative definitions of aggressive PCa (ie, Gleason score ≥8 or PSA at diagnosis >20 ng/ml), the overall patterns held, with higher PSA levels being associated with higher risk. If we apply the PPV of 21.9%, as was reported from the control arm of the Prostate Cancer Prevention Trial  in 5587 biopsied men with initial PSA ≤3.0 ng/ml (>55 yr of age) receiving biopsies upon indication (PSA >4.0 ng/ml or abnormal DRE) or at the end of a 7-yr study, a total of 3468 PCa diagnoses would be expected in 15 837 men. This conclusion indicates that 2474 PCa cases were not (yet) diagnosed in our cohort. The absence of PSA and clinical progression after 11 yr of follow-up is reassuring, but longer observation periods are needed to judge the natural history of these biopsy-detectable cancers. Our data show that in particular, subjects with PSA values close to the PSA cut-off have a high probability of being diagnosed with PCa in subsequent rounds. Previous studies confirm the association of PSA ranges with PCa by showing a gradual increase in the odds of cancer during follow-up with higher baseline PSA measurements  and .
In total, 3535 subjects died, resulting in an overall mortality rate of 21.4 per 1000 LY. The slightly higher OS in the group with an initial PSA level of 1.0–1.9 ng/ml compared to the PSA <1.0 group remains unexplained. The overall risk of PCa death in our study cohort was limited, with a death rate equal to 0.14 per 1000 LY, compared to 1.55 per 1000 LY (p
Our results are consistent with previous studies showing PSA to be a predictor of death from PCa, even at moderate PSA levels  and . Vickers et al.  reported comparable findings to ours from their cohort of unscreened men, showing that PCa deaths among men with initial PSA levels <1.0 ng/ml are rare. Results of the current study strengthen the justification of the use of PSA in risk stratification for screening purposes. The favourable outcome in men with PSA values <1.0 ng/ml supports the result of a previous study  to prolong the screening interval to 8 yr in this group. To contribute to decreasing PCa mortality by means of screening, a large number of PCa deaths would need to be prevented, which cannot be found in the low PSA ranges. More aggressive screening would lead to an unacceptable number needed to investigate (NNI) and number needed to treat (NNT). In line with this conclusion, a recent study  showed an NNI of 24 642 and an NNT of 724 for PSA values <2.0 ng/ml and in addition reveals excessive overdiagnosis and minor profit in PCa mortality of only 0.005 per 1000 LY between a screening cohort and a population-based cohort. Our results support the justification of a PSA threshold for screening based on current knowledge. Further effort should focus on the 2.0–2.9-ng/ml PSA range.
Obviously, mechanisms to selectively detect aggressive cancers are needed, because the increase in the PCa detection rate with higher baseline PSA values does not alter the fact that the greater part of cases does not harbour aggressive disease. Already now, risk-stratification instruments are available that avoid unnecessary biopsies and decrease overdiagnosis by incorporating clinically available risk modifiers and, once potentially indolent cancers are diagnosed, unnecessary treatments  and . In addition, further study on molecular and genetic markers selectively identifying aggressive disease is warranted.
This study provides insight into the natural history of PCa cases that remain undetected and the fate of screening- and interval-detected PCa in men with initial PSA values <3.0 ng/ml. It was shown that the risk of developing PCa and aggressive PCa as well as PCa-specific death in men participating in a screening trial with an initial PSA value <3.0 ng/ml significantly increases in patients with higher initial PSA values. Interval-detected cancers more often have aggressive characteristics and a substantial influence in causing disease-specific death, although the overall risk of PCa death in this cohort is low compared to men with initial PSA values ≥3.0 ng/ml (11-fold higher) and compared to the risk of other causes of death (150-fold higher). These results contribute to the risk stratification and individual management of men in PSA-based screening programmes.
Study concept and design: Schröder, Roobol, Bul.
Acquisition of data: Bul, Zhu.
Analysis and interpretation of data: Bul, van Leeuwen, Zhu, Schröder, Roobol.
Drafting of the manuscript: Bul, van Leeuwen, Zhu.
Critical revision of the manuscript for important intellectual content: Bul, Schröder, Roobol, van Leeuwen, Zhu.
Statistical analysis: Bul, van Leeuwen.
Obtaining funding: None.
Administrative, technical, or material support: van Leeuwen, Zhu.
Supervision: Schröder, Roobol.
Other (specify): None.
Financial disclosures: I certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.
Funding/Support and role of the sponsor: The ERSPC is supported by grants from the Dutch Cancer Society (KWF 94-869, 98-1657, 2002-277, 2006-3518); The Netherlands Organization for Health Research and Development (002822820, 22000106, 50-50110-98-311); the 6th Framework Program of the EU: P-Mark: LSHC-CT-2004-503011; Beckman-Coulter, Hybritech Inc.; and Europe against Cancer (SOC 95 35109, SOC 96 201869 05F02, SOC 97 201329, SOC 98 32241). The ERSPC received Erasmus MC and Ministry of Health institutional review board approval.
-  M.C. Wang, L.A. Valenzuela, G.P. Murphy, T.M. Chu. Purification of a human prostate specific antigen. Invest Urol. 1979;17:159-163
-  T.A. Stamey, N. Yang, A.R. Hay, J.E. McNeal, F.S. Freiha, E. Redwine. Prostate-specific antigen as a serum marker for adenocarcinoma of the prostate. N Engl J Med. 1987;317:909-916 Crossref.
-  W.J. Catalona, D.S. Smith, T.L. Ratliff, et al. Measurement of prostate-specific antigen in serum as a screening test for prostate cancer. N Engl J Med. 1991;324:1156-1161 Crossref.
-  J. Ferlay, D.M. Parkin, E. Steliarova-Foucher. Estimates of cancer incidence and mortality in Europe in 2008. Eur J Cancer. 2010;46:765-781 Crossref.
-  A. Jemal, R. Siegel, E. Ward, Y. Hao, J. Xu, M.J. Thun. Cancer statistics, 2009. CA Cancer J Clin. 2009;59:225-249 Crossref.
-  F.H. Schröder, J. Hugosson, M.J. Roobol, et al. Screening and prostate-cancer mortality in a randomized European study. N Engl J Med. 2009;360:1320-1328
-  G.L. Andriole, E.D. Crawford, R.L. Grubb III, et al. Mortality results from a randomized prostate-cancer screening trial. N Engl J Med. 2009;360:1310-1319 Crossref.
-  M.J. Roobol, M. Kerkhof, F.H. Schröder, et al. Prostate cancer mortality reduction by prostate-specific antigen–based screening adjusted for nonattendance and contamination in the European Randomised Study of Screening for Prostate Cancer (ERSPC). Eur Urol. 2009;56:584-591 Abstract, Full-text, PDF, Crossref.
-  I.M. Thompson, D.P. Ankerst, C. Chi, et al. Operating characteristics of prostate-specific antigen in men with an initial PSA level of 3.0 ng/ml or lower. JAMA. 2005;294:66-70 Crossref.
-  H.J. De Koning, J. Blom, J.W. Merkelbach, et al. Determining the cause of death in randomized screening trial(s) for prostate cancer. BJU Int. 2003;92(Suppl 2):71-78 Crossref.
-  M.J. Roobol, W.J. Kirkels, F.H. Schröder. Features and preliminary results of the Dutch Centre of the ERSPC (Rotterdam, The Netherlands). BJU Int. 2003;92(Suppl 2):48-54 Crossref.
-  A.V. D’Amico, R. Whittington, S.B. Malkowicz, et al. Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA. 1998;280:969-974 Crossref.
-  J. Hugosson, S. Carlsson, G. Aus, et al. Mortality results from the Goteborg randomised population-based prostate-cancer screening trial. Lancet Oncol. 2010;11:725-732 Crossref.
-  H. Lilja, D. Ulmert, T. Bjork, et al. Long-term prediction of prostate cancer up to 25 years before diagnosis of prostate cancer using prostate kallikreins measured at age 44 to 50 years. J Clin Oncol. 2007;25:431-436 Crossref.
-  P.H. Gann, C.H. Hennekens, M.J. Stampfer. A prospective evaluation of plasma prostate-specific antigen for detection of prostatic cancer. JAMA. 1995;273:289-294 Crossref.
-  M.J. Roobol, A. Grenabo, F.H. Schröder, J. Hugosson. Interval cancers in prostate cancer screening: comparing 2- and 4-year screening intervals in the European Randomized Study of Screening for Prostate Cancer, Gothenburg and Rotterdam. J Natl Cancer Inst. 2007;99:1296-1303 Crossref.
-  L.H. Kuller, A. Thomas, G. Grandits, J.D. Neaton, Multiple Risk Factor Intervention Trial Research Group. Elevated prostate-specific antigen levels up to 25 years prior to death from prostate cancer. Cancer Epidemiol Biomarkers Prev. 2004;13:373-377
-  D. Connolly, A. Black, A. Gavin, P.F. Keane, L.J. Murray. Baseline prostate-specific antigen level and risk of prostate cancer and prostate-specific mortality: diagnosis is dependent on the intensity of investigation. Cancer Epidemiol Biomarkers Prev. 2008;17:271-278 Crossref.
-  A.J. Vickers, A.M. Cronin, T. Bjork, et al. Prostate specific antigen concentration at age 60 and death or metastasis from prostate cancer: case-control study. BMJ. 2010;341:c4521 Crossref.
-  M.J. Roobol, D.W. Roobol, F.H. Schröder. Is additional testing necessary in men with prostate-specific antigen levels of 1.0 ng/mL or less in a population-based screening setting? (ERSPC, section Rotterdam). Urology. 2005;65:343-346 Crossref.
-  P.J. van Leeuwen, D. Connolly, T.L. Tammela, et al. Balancing the harms and benefits of early detection of prostate cancer. Cancer. 2010;116:4857-4865 Crossref.
-  M.J. Roobol, E.W. Steyerberg, R. Kranse, et al. A risk-based strategy improves prostate-specific antigen–driven detection of prostate cancer. Eur Urol. 2010;57:79-85 Abstract, Full-text, PDF, Crossref.
-  S. Perdonà, V. Cavadas, G. Di Lorenzo, et al. Prostate cancer detection in the “grey area” of prostate-specific antigen below 10 ng/ml: head-to-head comparison of the updated PCPT Calculator and Chun's nomogram, two risk estimators incorporating prostate cancer antigen 3. Eur Urol. 2011;59:81-87
© 2011 European Association of Urology, Published by Elsevier B.V.