Polygenicity, risk for prostate cancer, and the need for regular testing


A new study in the British Journal of Cancer suggests that knowledge of the presence of low or high numbers of known prostate cancer-related genes (polygenicity) “could” be helpful in determining which patients actually need regular testing for risk of prostate cancer. However, this sort of genetic analysis is complex, and so the more important question is likely to be whether polygenic analysis of this type is any more accurate at prediction of risk than a simple baseline PSA test for a man in his early 40s.

The new paper by Pashayan et al. is based on data defined by a polygenic risk score from nearly 5,000 men who participated in the Finnish patient group of the European Randomized Study of Screening for Prostate Cancer (ERSPC).

Based on the genotypes of 66 known prostate cancer-related loci, the authors calculated a polygenic risk score for a total of 4,967 Finnish participants in the ERSPC and then they stratified  all 72,072 men in the Finnish group as to whether their polygenic risk score was above or below a median value. They then estimated the percentage of screen-detected cancers that were probably “over-diagnosed” (i.e., never necessarily needing to be diagnosed) among the Finnish population and compared these data to the number of actually observed and expected cancers found by the screening trial.

They found that:

  • 74 percent of the screening-detected cancers in Finland occurred in men with polygenic risk above the median.
  • Sensitivity of the method was only 55 percent.
  • “Over-diagnosis” could be defined in 42 percent of all screen-detected prostate cancers, but
  • More than a third of those “over-diagnosed” cancers (37 percent) were found in men with higher than median polygenic risk.
  • Slightly more than half of those “over-diagnosed” cancers (58 percent) were found in men with lower than median polygenic risk.

The authors therefore concluded (accurately) that

Targeting screening to men at higher polygenic risk could reduce the proportion of cancers over-diagnosed.

The problem, of course, is that such a strategy would lead to “under-diagnosis” of potentially clinically significant prostate cancer in  > 40 percent of men with low polygenic risk who are actually found to have prostate cancer based on PSA testing and a subsequent biopsy.

We recognize that the whole question of what is implied by “over-diagnosis” comes with all its own issues. However, it would be nice for many men who are truly at low or very low risk for clinically significant prostate cancer to be able to know with a high degree of accuracy that they don’t need to be either tested or biopsied at all for risk of prostate cancer — if this is actually possible. The problem is that a test that seeks to offer patients that information need to have an accuracy of more like 95 percent than 50 percent. Polygenicity doesn’t seem to be offering that level of guidance … or at least, not based on the gene profiles assessed in the current study.

At the present time at least, a simple baseline PSA test appears to give at least as good and maybe a better set of data that correlates reasonably well with future risk for a diagnosis of prostate cancer.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: