On the use of prostate-specific antigen for screening of prostate cancer in European Randomised Study for Screening of Prostate Cancer

Prostate-specific antigen (PSA) has been the main drive for early detection of prostate cancer (PCa), including in population-based screening as in the European Randomised Study for Screening of Prostate Cancer (ERSPC). The specificity of PSA to indicate men with biopsy detectable prostate cancer can be improved by adding information obtained by new biomarkers, such as PSA isoforms. This improvement is needed to increase the efficacy of the screening procedure for the population-based as well as the individual screening. Various PSA isoforms, kallikreins and molecular markers have been validated in various cohorts from ERSPC of men with and without PCa in order to design the optimal diagnostic procedure for screening asymptomatic men. So far, most promising results have been obtained from the analysis of free PSA, proPSA, nicked PSA and hK2. The use of free PSA in addition to total PSA reduces the number of negative sextant biopsies at a PSA cut-off level of 3 ng/ml at initial screening with 30%, at the cost of losing 10% of detectable cancers that are predominantly well differentiated on histology. Further addition of PSA isoforms and hK2 only improve ROC curves in selected samples by a maximum of 5%. Molecular markers like PCA3 and TMPRSS2 in urine do not appear to be useful but they have been assessed insufficiently so far. The level of PSA at initial screening is highly predictive for the chance of being diagnosed with PCa later on in life. The changes in PSA over time after initial screening (like PSA-velocity and PSA-doubling time) are statistically different between men with detectable cancers versus those without (PSA-doubling time 5.1 versus 6.1 years), but this does not contribute significantly to population-based screening overall. Changes in specificity need to be related to a cost efficacy evaluation in the final analysis of ERSPC.

• View at Publisher
• View at Pubmed
• View at Scopus
• View at Web of Science