Comparison of screen detected and clinically diagnosed prostate cancer in the European randomized study of screening for prostate cancer, section rotterdam.

PURPOSE

This report provides a comparison of characteristics detected in the screening and control arms of the European Randomized study of Screening for Prostate Cancer, section Rotterdam.

MATERIALS AND METHODS

Between December 1993 and January 1999, 35,148 men 55 to 74 years old were randomized to European Randomized study of Screening for Prostate Cancer Rotterdam, including 17,635 in the screening arm and 17,513 in the control arm. Prostate specific antigen testing, digital rectal examination, transrectal ultrasound and sextant biopsies were offered to all participants in the screening arm according to 2 algorithms. All screening detected cancers and cancers found in the control arm were evaluated at the same cutoff point, that is January 1, 2003. To identify prostate cancer cases in the control arm yearly linkage was performed with the Rotterdam Cancer Registry database. Followup information was collected by chart review.

RESULTS

By January 1, 2003, 1,269 cancers were detected in the screening arm and 336 were detected in the control arm. A shift to more favorable clinical stages and histological grades on biopsy was seen in the screening arm of the trial. T1C and T2 cancers were 5.8 and 6.2 times more often diagnosed, respectively, in the screening arm than in the control arm of the trial. Only 4.6% of control arm cancers were found through opportunistic screening.

CONCLUSIONS

Although a favorable shift in prognostic factors was seen for the screening arm of the trial, these results do not provide evidence that prostate cancer screening has an effect on prostate cancer mortality.