Lesions predictive for prostate cancer in a screened population:: First and second screening round findings

BACKGROUND. We evaluated the incidence of prostate cancer, high-grade prostatic intraepithelial neoplasia (PIN) and lesions suspicious for prostate cancer (LSPC) in sextant biopsies in two subsequent screening rounds at a 4-year interval and their predictive value for subsequent prostate cancer detection. METHODS. In the Rotterdam section of the European Randomized Study of Screening for Prostate Cancer (ERSPC), 4,117 men underwent sextant biopsy in the 1st screening round. The 2nd round was performed at a 4-year interval and biopsies were taken in 1,840 men. RESULTS. The incidences of prostate cancer, LSPC and PIN in the 1st, respectively 2nd round were 24.6, resp. 19.9% (P = 0.001), 2.7, resp. 2.8% and 0.8, resp. 2.5% (P < 0.0001). Prostate cancer incidence after repeat biopsy for LSPC in the 1st, respectively 2nd round was 36.7 and 17.0%, and after repeat biopsy for PIN 13.3% in both rounds, respectively. Men with a benign biopsy in the 1st round had a significantly lower prostate cancer incidence in the 2nd round compared to men who did not undergo biopsy in the 1st round (10.7 vs. 22.7%, P < 0.0001). CONCLUSIONS. The decrease of prostate cancer detection in the 2nd round was associated with an increase in the incidence of PIN. Strikingly, LSPC diagnosed during the 1st round, but not during the 2nd round were predictive for prostate cancer, while isolated PIN was never predictive for prostate cancer. PIN shoud not be an indication for repeat biopsy in a screening population. Importantly a 1st round benign biopsy outcome proved to be a negative predictor of subsequent prostate cancer detection. (C) 2004 Wiley-Liss, Inc.