Letter to the Editor
Prostate-specific antigen screening for prostate cancer: Why so much controversy?
Article first published online: 15 APR 2013
© 2013 The Japanese Urological Association
International Journal of Urology
Volume 20, Issue 11, page 1148, November 2013
How to Cite
Labrie, F. (2013), Prostate-specific antigen screening for prostate cancer: Why so much controversy?. International Journal of Urology, 20: 1148. doi: 10.1111/iju.12162
- Issue published online: 29 OCT 2013
- Article first published online: 15 APR 2013
As prostate cancer remains the second leading cause of cancer deaths in men with 238 590 new cases and 29 720 deaths estimated for 2013 in the USA alone, it seems important to further examine the data responsible for the surprising recommendations against prostate cancer screening. As prostate cancer progresses with no symptoms, these negative opinions need to be reconciled with the recognized fact that without screening, the majority of men will reach the advanced metastatic stage before being diagnosed, thus losing the possibility of a cure and regressing 40 years in the approach to prostate cancer diagnosis and treament.
An update at 11 years of follow up of the European Randomized Study of Screening for Prostate Cancer (ERSPC) shows an overall 21% reduction in the rate of prostate cancer deaths, with a relative risk reduction of 38% during years 10 and 11. In the Quebec study, at a median of 7.93 years of follow up, a 62% decrease in prostate cancer deaths was observed.
Contrary to the European and Quebec studies, the USA Prostate, Lung, Colon and Ovarian Cancer (PLCO) study reported no influence of screening on deaths from prostate cancer. Examination of the PLCO trial, however, shows that 79% of men in the so-called “control” group at year 5 had a previous prostate-specific antigen (PSA) test compared with 91% in the screened group. In fact, just 13% of men reported never having a PSA or digital rectal examination test in the intended “control” group. In short, it has been estimated that the PLCO trial had up to 85% of men screened in the “control” arm where no man should have been screened.
With such a high rate of contamination of men screened in the “control” group, the PLCO study did not have the statistical power to provide any meaningful information. In fact, it was already too late in 1993 to start a screening trial in the USA, as too many men had already been screened.
It can be added that the important benefits of screening observed in the European study were achieved using a less than optimal 4-year PSA screening interval, which misses a significant number of cancers, whereas the Quebec study used the 1-year interval, thus potentially explaining the 62% benefit observed in favor of screening in the Quebec trial compared with 38% in years 10 and 11 in the European study. In fact, only men having a PSA below 1.0 ng/mL can be screened every 4 years with no significant negative impact on the time of cancer diagnosis.
From the aforementioned facts, it seems reasonable to suggest that the USA study cannot be used to support any recommendation about screening. Recommendations about screening should only be based on data obtained in clinical trials that have sufficient statistical power, and are free of serious flaws and bias; the absence of a true control group being a fundamental issue. Although some criticism can be expressed, the best conditions for a screening evaluation have been those of the European and Quebec trials, which both show a very important reduction of prostate cancer deaths by screening.