PSA and the Detection of Prostate Cancer After 2005. Part I

PSA has been shown to be a powerful tool to detect prostate cancer in men at risk due to their age. Catalona et al. [1] used a PSA cut-off value of 4 ng/mL in a group of 1653 men in age groups at risk as a biopsy indication. 37 cancers (2.3%) were detected. 16 of the 37 cancers would have been missed by rectal examination alone. With increasing use of PSA it then emerged that the prostate cancers detected in excess of rectal examination had a more favourable distribution of prognostic factors. This was again shown early on in a prospective setting by Catalona et al. [2]. Recent developments have cast doubt on the use of the traditional PSA cut-off of 4.0 ng/mL and on the value of PSA levels in general and specifically in heavily pre-screened populations. It seems that recent findings, which will be subject to review in this paper, have made the role of PSA in diagnosing prostate cancer more uncertain than ever before. While lowering PSA cut-off levels leads to a higher detection rate of prostate cancer, it also leads to an increase of the diagnosis of cancers which might otherwise never bother their carrier (overdiagnosis). In addition, potentially aggressive cancers as defined by Gleason scores above 7 cannot be identified by appropriate PSA cut-off levels. Even with PSA values below 3.0 ng/mL 10-15% of cancers identified have been shown to have aggressive features. The high rate of overdiagnosis makes strategies desirable, which identify aggressive but still curable cancers with an acceptable accuracy. This unfortunately is still impossible. Potentially valid approaches, which include the use of PSA kinetics over time, are in the process of being evaluated. The best way of making use of PSA in the future for diagnosing prostate cancer and its role in differentiating between aggressive and less aggressive tumours seems uncertain at this time. Further research in this field is urgently needed and directions are outlined. For the time being, arbitrary cut-off levels will remain the best option even in repeat screening.

• View at Publisher
• View at Scopus