A re-analysis of data from the Antwerp (Belgium) section of the European Randomized Study of Screening for Prostate Cancer (ERSPC) has offered us some further insight into the frequency of PSA testing necessary to identify less and more aggressive forms of prostate cancer.
Nelen et al. have compiled data on the number and characteristics of so-called “interval cancers” in men in the screening arm of the Antwerp section of ERSPC who were between 55 and 65 years of age at initial randomization and who participated in all of the screening rounds they were invited to attend. The total number of men in the screening arm of the Antwerp section was about 1,600. The screening interval in the Antwerp section of the ERSCP lasted for 6 years on average.
It is important, in understanding this study to appreciate exactly what is meant by an “interval cancer.” The authors defined “interval cancers” as:
- Cases of prostate cancer diagnosed during the screening interval but not detected by screening and/or
- Cases of prostate cancer diagnosed as a consequence of a positive screening test if diagnosis through biopsy occurred more than 1 year after that screening test.
Aggressive types of interval cancers were defined as interval cancers with at least one of the following characteristics:
- Clinical stage M1 or N1
- Gleason score of 8 to 10
- World Health Organisation (WHO) score of 3.
The results of this study can be summarized as follows:
- The 10-year cumulative incidence of interval cancers was 3.0 percent (n = 50).
- The cumulative incidence of aggressive interval cancers was 0.5 percent (n = 8).
- During the first screening interval
- 36 interval cancers were detected.
- 20/36 interval cancers (55.6 percent) were detected more than 4 years after the initial screening.
- 5/36 interval cancers (13.9 percent) were classified as aggressive.
- All aggressive interval cancers emerged > 4 years after initial screening.
The authors conclude only that the occurrence of interval cancers in this study was higher than in the ERSPC centres that used a shorter screening interval and that aggressive interval cancers started to emerge at least 4 years after initial screening.
We have previously suggested that — at least among men who have no well-defined risk for prostate cancer — PSA testing for risk of prostate cancer may be needed only at 5-year intervals to identify clinically signioficant forms of prostate cancer. It would be inappropriate to suggest that the data from this study specifically supports that suggestion. However, it does at least appear to give some further credence to the hypothesis that an extended period between PSA tests does not place most men at high risk for a diagnosis of clinically significant disease.
We would hope that additional study might be able to clarify whether 3, 4 or 5 years was the most appropriate period between PSA tests necessary to monitor risk in men over 50 years of age with no known risk factors for prostate cancer. We are increasingly convinced that annual PSA tests are unnecessary for the detection of risk among the majority of men of 40+ years of age. Annual testing may, however, be wise for men with known risks for prostate cancer.