BACKGROUND. The identification of predictors for prostate biopsy outcome at two screening rounds using a PSA ≥3.0 ng/ml as biopsy indication. MATERIALS AND METHODS. We compared predictors by means of descriptive statistics and logistic regression analysis in men (55-75 years) biopsied in either the 1st or 2nd screening round of ERSPC Rotterdam (interval 4 years). RESULTS. Positive predictors for biopsy outcome in both screening rounds were an increased PSA level in the absence of a previous negative biopsy (PrevNB), DRE and TRUS suspicious and a positive family history (PFH). A higher than median prostate volume was a consistent negative predictor. Having had a PrevNB at initial screening strongly reduced the chance of cancer detection at repeat screening and in addition canceled the predictive potential of PSA. CONCLUSION. If "detecting prostate cancer efficiently" were the aim, this study indicates that a "PSA only based biopsy threshold" may be replaced by another criterion incorporating, e.g., DRE, TRUS and prostate volume in men who were biopsied in the preceding 4 year interval.

Logistic regression analysis, Prostate biopsy, Prostate cancer, PSA, Screening
dx.doi.org/10.1002/pros.20352, hdl.handle.net/1765/68194
The Prostate
Department of Urology

Roobol-Bouts, M.J, Schröder, F.H, & Kranse, R. (2006). A comparison of first and repeat (four years later) prostate cancer screening in a randomized cohort of symptomatic men aged 55-75 years using a biopsy indication of 3.0 ng/ml (results of ERSPC, Rotterdam). The Prostate, 66(6), 604–612. doi:10.1002/pros.20352