Pathologic Features of Prostate Cancer Found at Population-Based Screening with a Four-Year Interval

Published in: Journal of the National Cancer Institute, v. 93, no. 15, Aug. 1, 2001, p.1153-1158

Posted on on January 01, 2001

by Robert F. Hoedemaeker, Harry H van der Kwast, Rob Boer, Harry J. de Koning, Monique Roobol, Andre N. Vis, Fritz H. Schroder

BACKGROUND: The currently recommended frequency for prostate-specific antigen (PSA) screening tests for prostate cancer is 1 year, but the optimal screening interval is not known. Our goal was to determine if a longer interval would compromise the detection of curable prostate cancer. METHODS: A cohort of 4491 men aged 55-75 years, all participants in the Rotterdam section of the European Randomized Study of (population-based) Screening for Prostate Cancer, were invited to participate in an initial PSA screening. Men who received that screening were invited for a second screen 4 years later. Pathology findings from needle biopsy cores were compared for men in both rounds. Statistical tests were two-sided. RESULTS: A total of 4133 men were screened in the first round (the prevalence screen), and 2385 were screened in the second round. The median amount of cancer in needle biopsy sets was 7.0 mm (95% confidence interval [CI] = 5.4 mm to 8.6 mm) in the first round and 4.1 mm (95% CI = 2.6 mm to 5.6 mm) in the second round (P =.001). Thirty-six percent of the adenocarcinomas detected in the first round but only 16% of those detected in the second round had a Gleason score of 7 or higher (mean difference = 20% [95% CI = 10% to 30%]; P<.001). Whereas 25% of the adenocarcinomas detected in the first round had adverse prognostic features, only 6% of those detected in the second round did (mean difference = 19% [95% CI = 11% to 26%]; P<.001). Baseline PSA values were predictive for the amount of tumor in biopsies in men with cancer in the first round but not for that in the second round. CONCLUSION: Most large prostate cancers with high serum PSA levels were effectively detected in a prevalence screen. In this population, a screening interval of 4 years appears to be short enough to constrain the development of large tumors, although it is inconclusive whether this will result in a survival benefit.

This report is part of the RAND Corporation External publication series. Many RAND studies are published in peer-reviewed scholarly journals, as chapters in commercial books, or as documents published by other organizations.

The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.